valknar/llmx
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
a67a67f3258fc21e147b6786a143fe3e15e6d5ba
llmx/codex-rs/Cargo.lock

5423 lines
131 KiB
Plaintext
Raw Normal View History

feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
# This file is automatically @generated by Cargo.
# It is not intended for manual editing.
version = 4
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "Inflector"
version = "0.11.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fe438c63458706e03479442743baae6c88256498e6431708f6dfc520a26515d3"
dependencies = [
"lazy_static",
"regex",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "addr2line"
version = "0.21.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8a30b2e23b9e17a9f90641c7ab1549cd9b44f296d3ccbf309d2863cfe398a0cb"
dependencies = [
"gimli",
]
[[package]]
name = "adler"
version = "1.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f26201604c87b1e01bd3d98f8d5d9a8fcbb815e8cedb41ffccbeb4bf593a35fe"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "adler2"
version = "2.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "512761e0bb2578dd7380c6baaa0f4ce03e84f95e960231d1dec8bf4d7d6e2627"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "ahash"
version = "0.8.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e89da841a80418a9b391ebaea17f5c112ffaaa96f621d2c285b5174da76b9011"
dependencies = [
"cfg-if",
"once_cell",
"version_check",
"zerocopy 0.7.35",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "aho-corasick"
version = "1.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8e60d3430d3a69478ad0993f19238d2df97c507009a52b3c10addcd7f6bcb916"
dependencies = [
"memchr",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "aligned-vec"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4aa90d7ce82d4be67b64039a3d588d38dbcc6736577de4a847025ce5b0c468d1"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "allocative"
version = "0.3.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8fac2ce611db8b8cee9b2aa886ca03c924e9da5e5295d0dbd0526e5d0b0710f7"
dependencies = [
"allocative_derive",
"bumpalo",
"ctor",
"hashbrown 0.14.5",
"num-bigint",
]
[[package]]
name = "allocative_derive"
version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fe233a377643e0fc1a56421d7c90acdec45c291b30345eb9f08e8d0ddce5a4ab"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "allocator-api2"
version = "0.2.21"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "683d7910e743518b0e34f1186f92494becacb047c7b6bf616c96772180fef923"
[[package]]
name = "android-tzdata"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e999941b234f3131b00bc13c22d06e8c5ff726d1b6318ac7eb276997bbb4fef0"
[[package]]
name = "android_system_properties"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "819e7219dbd41043ac279b19830f2efc897156490d7fd6ea916720117ee66311"
dependencies = [
"libc",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "annotate-snippets"
version = "0.9.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ccaf7e9dfbb6ab22c82e473cd1a8a7bd313c19a5b7e40970f3d89ef5a5c9e81e"
dependencies = [
"unicode-width 0.1.14",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "ansi-to-tui"
version = "7.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "67555e1f1ece39d737e28c8a017721287753af3f93225e4a445b29ccb0f5912c"
dependencies = [
"nom",
"ratatui",
"simdutf8",
"smallvec",
"thiserror 1.0.69",
]
[[package]]
name = "anstream"
version = "0.6.18"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8acc5369981196006228e28809f761875c0327210a891e941f4c683b3a99529b"
dependencies = [
"anstyle",
"anstyle-parse",
"anstyle-query",
"anstyle-wincon",
"colorchoice",
"is_terminal_polyfill",
"utf8parse",
]
[[package]]
name = "anstyle"
version = "1.0.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "55cc3b69f167a1ef2e161439aa98aed94e6028e5f9a59be9a6ffb47aef1651f9"
[[package]]
name = "anstyle-parse"
version = "0.2.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3b2d16507662817a6a20a9ea92df6652ee4f94f914589377d69f3b21bc5798a9"
dependencies = [
"utf8parse",
]
[[package]]
name = "anstyle-query"
version = "1.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "79947af37f4177cfead1110013d678905c37501914fba0efea834c3fe9a8d60c"
dependencies = [
"windows-sys 0.59.0",
]
[[package]]
name = "anstyle-wincon"
version = "3.0.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ca3534e77181a9cc07539ad51f2141fe32f6c3ffd4df76db8ad92346b003ae4e"
dependencies = [
"anstyle",
"once_cell",
"windows-sys 0.59.0",
]
[[package]]
name = "anyhow"
version = "1.0.98"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e16d2d3311acee920a9eb8d33b8cbc1787ce4a264e85f964c2404b969bdcd487"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "arbitrary"
version = "1.4.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dde20b3d026af13f561bdd0f15edf01fc734f0dafcedbaf42bba506a9517f223"
[[package]]
name = "arg_enum_proc_macro"
version = "0.3.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0ae92a5119aa49cdbcf6b9f893fe4e1d98b04ccbf82ee0584ad948a44a734dea"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
[[package]]
name = "arrayvec"
version = "0.7.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7c02d123df017efcdfbd739ef81735b36c5ba83ec3c59c80a9d7ecc718f92e50"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "ascii-canvas"
version = "3.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8824ecca2e851cec16968d54a01dd372ef8f95b244fb84b84e70128be347c3c6"
dependencies = [
"term",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "assert-json-diff"
version = "2.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "47e4f2b81832e72834d7518d8487a0396a28cc408186a2e8854c0f98011faf12"
dependencies = [
"serde",
"serde_json",
]
[[package]]
name = "assert_cmd"
version = "2.0.17"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2bd389a4b2970a01282ee455294913c0a43724daedcd1a24c3eb0ec1c1320b66"
dependencies = [
"anstyle",
"bstr",
"doc-comment",
"libc",
"predicates",
"predicates-core",
"predicates-tree",
"wait-timeout",
]
[[package]]
name = "async-channel"
version = "2.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "89b47800b0be77592da0afd425cc03468052844aff33b84e33cc696f64e77b6a"
dependencies = [
"concurrent-queue",
"event-listener-strategy",
"futures-core",
"pin-project-lite",
]
[[package]]
name = "async-trait"
version = "0.1.88"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e539d3fca749fcee5236ab05e93a52867dd549cc157c8cb7f99595f3cedffdb5"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "atomic-waker"
version = "1.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1505bd5d3d116872e7271a6d4e16d81d0c8570876c8de68093a09ac269d8aac0"
[[package]]
name = "autocfg"
version = "1.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ace50bade8e6234aa140d9a2f552bbee1db4d353f69b8217bc503490fc1a9f26"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "av1-grain"
version = "0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4f3efb2ca85bc610acfa917b5aaa36f3fcbebed5b3182d7f877b02531c4b80c8"
dependencies = [
"anyhow",
"arrayvec",
"log",
"nom",
"num-rational",
"v_frame",
]
[[package]]
name = "avif-serialize"
version = "0.8.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "98922d6a4cfbcb08820c69d8eeccc05bb1f29bfa06b4f5b1dbfe9a868bd7608e"
dependencies = [
"arrayvec",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "backtrace"
version = "0.3.71"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "26b05800d2e817c8b3b4b54abd461726265fa9789ae34330622f2db9ee696f9d"
dependencies = [
"addr2line",
"cc",
"cfg-if",
"libc",
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
"miniz_oxide 0.7.4",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"object",
"rustc-demangle",
]
[[package]]
name = "base64"
version = "0.21.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9d297deb1925b89f2ccc13d7635fa0714f12c87adce1c75356b39ca9b7178567"
[[package]]
name = "base64"
version = "0.22.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "72b3254f16251a8381aa12e40e3c4d2f0199f8c6508fbecb9d91f575e0fbb8c6"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "beef"
version = "0.5.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3a8241f3ebb85c056b509d4327ad0358fbbba6ffb340bf388f26350aeda225b1"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "bincode"
version = "1.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b1f45e9417d87227c7a56d22e471c6206462cba514c7590c09aff4cf6d1ddcad"
dependencies = [
"serde",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "bit-set"
version = "0.5.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0700ddab506f33b20a03b13996eccd309a48e5ff77d0d95926aa0210fb4e95f1"
dependencies = [
"bit-vec",
]
[[package]]
name = "bit-vec"
version = "0.6.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "349f9b6a179ed607305526ca489b34ad0a41aed5f7980fa90eb03160b69598fb"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "bit_field"
version = "0.10.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dc827186963e592360843fb5ba4b973e145841266c1357f7180c43526f2e5b61"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "bitflags"
version = "1.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "bitflags"
version = "2.9.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5c8214115b7bf84099f1309324e63141d4c5d7cc26862f97a0a857dbefe165bd"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "bitstream-io"
version = "2.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6099cdc01846bc367c4e7dd630dc5966dccf36b652fae7a74e17b640411a91b2"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "bstr"
version = "1.12.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "234113d19d0d7d613b40e86fb654acf958910802bcceab913a4f9e7cda03b1a4"
dependencies = [
"memchr",
"regex-automata 0.4.9",
"serde",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "built"
version = "0.7.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56ed6191a7e78c36abdb16ab65341eefd73d64d303fffccdbb00d51e4205967b"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "bumpalo"
version = "3.17.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1628fb46dfa0b37568d12e5edd512553eccf6a22a78e8bde00bb4aed84d5bdbf"
[[package]]
name = "bytecount"
version = "0.6.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5ce89b21cab1437276d2650d57e971f9d548a2d9037cc231abdc0562b97498ce"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "bytemuck"
version = "1.23.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9134a6ef01ce4b366b50689c94f82c14bc72bc5d0386829828a2e2752ef7958c"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "byteorder"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1fd0f2584146f6f2ef48085050886acf353beff7305ebd1ae69500e27c67f64b"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "byteorder-lite"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8f1fe948ff07f4bd06c30984e69f5b4899c516a3ef74f34df92a2df2ab535495"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "bytes"
version = "1.10.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d71b6127be86fdcfddb610f7182ac57211d4b18a3e9c82eb2d17662f2227ad6a"
[[package]]
name = "cassowary"
version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "df8670b8c7b9dae1793364eafadf7239c40d669904660c5960d74cfd80b46a53"
[[package]]
name = "castaway"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0abae9be0aaf9ea96a3b1b8b1b55c602ca751eba1b1500220cea4ecbafe7c0d5"
dependencies = [
"rustversion",
]
[[package]]
name = "cc"
version = "1.2.19"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8e3a13707ac958681c13b39b458c073d0d9bc8a22cb1b2f4c8e55eb72c13f362"
dependencies = [
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"jobserver",
"libc",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"shlex",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "cfg-expr"
version = "0.15.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d067ad48b8650848b989a59a86c6c36a995d02d2bf778d45c3c5d57bc2718f02"
dependencies = [
"smallvec",
"target-lexicon",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "cfg-if"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "cfg_aliases"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fd16c4719339c4530435d38e511904438d07cce7950afa3718a84ac36c10e89e"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "chrono"
version = "0.4.40"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1a7964611d71df112cb1730f2ee67324fcf4d0fc6606acbbe9bfe06df124637c"
dependencies = [
"android-tzdata",
"iana-time-zone",
"js-sys",
"num-traits",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"serde",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"wasm-bindgen",
"windows-link",
]
[[package]]
name = "clap"
version = "4.5.37"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "eccb054f56cbd38340b380d4a8e69ef1f02f1af43db2f0cc817a4774d80ae071"
dependencies = [
"clap_builder",
"clap_derive",
]
[[package]]
name = "clap_builder"
version = "4.5.37"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "efd9466fac8543255d3b1fcad4762c5e116ffe808c8a3043d4263cd4fd4862a2"
dependencies = [
"anstream",
"anstyle",
"clap_lex",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"strsim 0.11.1",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"terminal_size",
]
[[package]]
name = "clap_derive"
version = "4.5.32"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "09176aae279615badda0765c0c0b3f6ed53f4709118af73cf4655d85d1530cd7"
dependencies = [
"heck",
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "clap_lex"
version = "0.7.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f46ad14479a25103f283c0f10005961cf086d8dc42205bb44c46ac563475dca6"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "clipboard-win"
version = "5.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "15efe7a882b08f34e38556b14f2fb3daa98769d06c7f0c1b076dfd0d983bc892"
dependencies = [
"error-code",
]
[[package]]
name = "cmp_any"
version = "0.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e9b18233253483ce2f65329a24072ec414db782531bdbb7d0bbc4bd2ce6b7e21"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "codex-ansi-escape"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
dependencies = [
"ansi-to-tui",
"ratatui",
"tracing",
]
[[package]]
name = "codex-apply-patch"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
dependencies = [
"anyhow",
"pretty_assertions",
"serde_json",
"similar",
"tempfile",
"thiserror 2.0.12",
"tree-sitter",
"tree-sitter-bash",
]
[[package]]
name = "codex-cli"
chore: script to create a Rust release (#759) For now, keep things simple such that we never update the `version` in the `Cargo.toml` for the workspace root on the `main` branch. Instead, create a new branch for a release, push one commit that updates the `version`, and then tag that branch to kick off a release. To test, I ran this script and created this release job: https://github.com/openai/codex/actions/runs/14762580641
2025-04-30 12:39:03 -07:00
version = "0.0.0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
dependencies = [
"anyhow",
"clap",
chore: introduce codex-common crate (#843) I started this PR because I wanted to share the `format_duration()` utility function in `codex-rs/exec/src/event_processor.rs` with the TUI. The question was: where to put it? `core` should have as few dependencies as possible, so moving it there would introduce a dependency on `chrono`, which seemed undesirable. `core` already had this `cli` feature to deal with a similar situation around sharing common utility functions, so I decided to: * make `core` feature-free * introduce `common` * `common` can have as many "special interest" features as it needs, each of which can declare their own deps * the first two features of common are `cli` and `elapsed` In practice, this meant updating a number of `Cargo.toml` files, replacing this line: ```toml codex-core = { path = "../core", features = ["cli"] } ``` with these: ```toml codex-core = { path = "../core" } codex-common = { path = "../common", features = ["cli"] } ``` Moving `format_duration()` into its own file gave it some "breathing room" to add a unit test, so I had Codex generate some tests and new support for durations over 1 minute.
2025-05-06 17:38:56 -07:00
"codex-common",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"codex-core",
"codex-exec",
fix: overhaul how we spawn commands under seccomp/landlock on Linux (#1086) Historically, we spawned the Seatbelt and Landlock sandboxes in substantially different ways: For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy specified as an arg followed by the original command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec.rs#L147-L219 For **Landlock/Seccomp**, we would do `tokio::runtime::Builder::new_current_thread()`, _invoke Landlock/Seccomp APIs to modify the permissions of that new thread_, and then spawn the command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec_linux.rs#L28-L49 While it is neat that Landlock/Seccomp supports applying a policy to only one thread without having to apply it to the entire process, it requires us to maintain two different codepaths and is a bit harder to reason about. The tipping point was https://github.com/openai/codex/pull/1061, in which we had to start building up the `env` in an unexpected way for the existing Landlock/Seccomp approach to continue to work. This PR overhauls things so that we do similar things for Mac and Linux. It turned out that we were already building our own "helper binary" comparable to Mac's `sandbox-exec` as part of the `cli` crate: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/cli/Cargo.toml#L10-L12 We originally created this to build a small binary to include with the Node.js version of the Codex CLI to provide support for Linux sandboxing. Though the sticky bit is that, at this point, we still want to deploy the Rust version of Codex as a single, standalone binary rather than a CLI and a supporting sandboxing binary. To satisfy this goal, we use "the arg0 trick," in which we: * use `std::env::current_exe()` to get the path to the CLI that is currently running * use the CLI as the `program` for the `Command` * set `"codex-linux-sandbox"` as arg0 for the `Command` A CLI that supports sandboxing should check arg0 at the start of the program. If it is `"codex-linux-sandbox"`, it must invoke `codex_linux_sandbox::run_main()`, which runs the CLI as if it were `codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn the original command, so do _replace_ the process rather than spawn a subprocess. Incidentally, we do this before starting the Tokio runtime, so the process should only have one thread when `execvp(3)` is called. Because the `core` crate that needs to spawn the Linux sandboxing is not a CLI in its own right, this means that every CLI that includes `core` and relies on this behavior has to (1) implement it and (2) provide the path to the sandboxing executable. While the path is almost always `std::env::current_exe()`, we needed to make this configurable for integration tests, so `Config` now has a `codex_linux_sandbox_exe: Option<PathBuf>` property to facilitate threading this through, introduced in https://github.com/openai/codex/pull/1089. This common pattern is now captured in `codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs` functions that should use it have been updated as part of this PR. The `codex-linux-sandbox` crate added to the Cargo workspace as part of this PR now has the bulk of the Landlock/Seccomp logic, which makes `core` a bit simpler. Indeed, `core/src/exec_linux.rs` and `core/src/landlock.rs` were removed/ported as part of this PR. I also moved the unit tests for this code into an integration test, `linux-sandbox/tests/landlock.rs`, in which I use `env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for `codex_linux_sandbox_exe` since `std::env::current_exe()` is not appropriate in that case.
2025-05-23 11:37:07 -07:00
"codex-linux-sandbox",
feat: add mcp subcommand to CLI to run Codex as an MCP server (#934) Previously, running Codex as an MCP server required a standalone binary in our Cargo workspace, but this PR makes it available as a subcommand (`mcp`) of the main CLI. Ran this with: ``` RUST_LOG=debug npx @modelcontextprotocol/inspector cargo run --bin codex -- mcp ``` and verified it worked as expected in the inspector at `http://127.0.0.1:6274/`.
2025-05-14 13:15:41 -07:00
"codex-mcp-server",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"codex-tui",
"serde_json",
"tokio",
"tracing",
"tracing-subscriber",
]
chore: introduce codex-common crate (#843) I started this PR because I wanted to share the `format_duration()` utility function in `codex-rs/exec/src/event_processor.rs` with the TUI. The question was: where to put it? `core` should have as few dependencies as possible, so moving it there would introduce a dependency on `chrono`, which seemed undesirable. `core` already had this `cli` feature to deal with a similar situation around sharing common utility functions, so I decided to: * make `core` feature-free * introduce `common` * `common` can have as many "special interest" features as it needs, each of which can declare their own deps * the first two features of common are `cli` and `elapsed` In practice, this meant updating a number of `Cargo.toml` files, replacing this line: ```toml codex-core = { path = "../core", features = ["cli"] } ``` with these: ```toml codex-core = { path = "../core" } codex-common = { path = "../common", features = ["cli"] } ``` Moving `format_duration()` into its own file gave it some "breathing room" to add a unit test, so I had Codex generate some tests and new support for durations over 1 minute.
2025-05-06 17:38:56 -07:00
[[package]]
name = "codex-common"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
chore: introduce codex-common crate (#843) I started this PR because I wanted to share the `format_duration()` utility function in `codex-rs/exec/src/event_processor.rs` with the TUI. The question was: where to put it? `core` should have as few dependencies as possible, so moving it there would introduce a dependency on `chrono`, which seemed undesirable. `core` already had this `cli` feature to deal with a similar situation around sharing common utility functions, so I decided to: * make `core` feature-free * introduce `common` * `common` can have as many "special interest" features as it needs, each of which can declare their own deps * the first two features of common are `cli` and `elapsed` In practice, this meant updating a number of `Cargo.toml` files, replacing this line: ```toml codex-core = { path = "../core", features = ["cli"] } ``` with these: ```toml codex-core = { path = "../core" } codex-common = { path = "../common", features = ["cli"] } ``` Moving `format_duration()` into its own file gave it some "breathing room" to add a unit test, so I had Codex generate some tests and new support for durations over 1 minute.
2025-05-06 17:38:56 -07:00
dependencies = [
"clap",
"codex-core",
feat: add support for -c/--config to override individual config items (#1137) This PR introduces support for `-c`/`--config` so users can override individual config values on the command line using `--config name=value`. Example: ``` codex --config model=o4-mini ``` Making it possible to set arbitrary config values on the command line results in a more flexible configuration scheme and makes it easier to provide single-line examples that can be copy-pasted from documentation. Effectively, it means there are four levels of configuration for some values: - Default value (e.g., `model` currently defaults to `o4-mini`) - Value in `config.toml` (e.g., user could override the default to be `model = "o3"` in their `config.toml`) - Specifying `-c` or `--config` to override `model` (e.g., user can include `-c model=o3` in their list of args to Codex) - If available, a config-specific flag can be used, which takes precedence over `-c` (e.g., user can specify `--model o3` in their list of args to Codex) Now that it is possible to specify anything that could be configured in `config.toml` on the command line using `-c`, we do not need to have a custom flag for every possible config option (which can clutter the output of `--help`). To that end, as part of this PR, we drop support for the `--disable-response-storage` flag, as users can now specify `-c disable_response_storage=true` to get the equivalent functionality. Under the hood, this works by loading the `config.toml` into a `toml::Value`. Then for each `key=value`, we create a small synthetic TOML file with `value` so that we can run the TOML parser to get the equivalent `toml::Value`. We then parse `key` to determine the point in the original `toml::Value` to do the insert/replace. Once all of the overrides from `-c` args have been applied, the `toml::Value` is deserialized into a `ConfigToml` and then the `ConfigOverrides` are applied, as before.
2025-05-27 23:11:44 -07:00
"serde",
"toml",
chore: introduce codex-common crate (#843) I started this PR because I wanted to share the `format_duration()` utility function in `codex-rs/exec/src/event_processor.rs` with the TUI. The question was: where to put it? `core` should have as few dependencies as possible, so moving it there would introduce a dependency on `chrono`, which seemed undesirable. `core` already had this `cli` feature to deal with a similar situation around sharing common utility functions, so I decided to: * make `core` feature-free * introduce `common` * `common` can have as many "special interest" features as it needs, each of which can declare their own deps * the first two features of common are `cli` and `elapsed` In practice, this meant updating a number of `Cargo.toml` files, replacing this line: ```toml codex-core = { path = "../core", features = ["cli"] } ``` with these: ```toml codex-core = { path = "../core" } codex-common = { path = "../common", features = ["cli"] } ``` Moving `format_duration()` into its own file gave it some "breathing room" to add a unit test, so I had Codex generate some tests and new support for durations over 1 minute.
2025-05-06 17:38:56 -07:00
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "codex-core"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
dependencies = [
"anyhow",
"assert_cmd",
"async-channel",
"base64 0.21.7",
"bytes",
"codex-apply-patch",
feat: support mcp_servers in config.toml (#829) This adds initial support for MCP servers in the style of Claude Desktop and Cursor. Note this PR is the bare minimum to get things working end to end: all configured MCP servers are launched every time Codex is run, there is no recovery for MCP servers that crash, etc. (Also, I took some shortcuts to change some fields of `Session` to be `pub(crate)`, which also means there are circular deps between `codex.rs` and `mcp_tool_call.rs`, but I will clean that up in a subsequent PR.) `codex-rs/README.md` is updated as part of this PR to explain how to use this feature. There is a bit of plumbing to route the new settings from `Config` to the business logic in `codex.rs`. The most significant chunks for new code are in `mcp_connection_manager.rs` (which defines the `McpConnectionManager` struct) and `mcp_tool_call.rs`, which is responsible for tool calls. This PR also introduces new `McpToolCallBegin` and `McpToolCallEnd` event types to the protocol, but does not add any handlers for them. (See https://github.com/openai/codex/pull/836 for initial usage.) To test, I added the following to my `~/.codex/config.toml`: ```toml # Local build of https://github.com/hideya/mcp-server-weather-js [mcp_servers.weather] command = "/Users/mbolin/code/mcp-server-weather-js/dist/index.js" args = [] ``` And then I ran the following: ``` codex-rs$ cargo run --bin codex exec 'what is the weather in san francisco' [2025-05-06T22:40:05] Task started: 1 [2025-05-06T22:40:18] Agent message: Here’s the latest National Weather Service forecast for San Francisco (downtown, near 37.77° N, 122.42° W): This Afternoon (Tue): • Sunny, high near 69 °F • West-southwest wind around 12 mph Tonight: • Partly cloudy, low around 52 °F • SW wind 7–10 mph ... ``` Note that Codex itself is not able to make network calls, so it would not normally be able to get live weather information like this. However, the weather MCP is [currently] not run under the Codex sandbox, so it is able to hit `api.weather.gov` and fetch current weather information. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/829). * #836 * __->__ #829
2025-05-06 15:47:59 -07:00
"codex-mcp-client",
fix: remove dependency on expanduser crate (#667) In putting up https://github.com/openai/codex/pull/665, I discovered that the `expanduser` crate does not compile on Windows. Looking into it, we do not seem to need it because we were only using it with a value that was passed in via a command-line flag, so the shell expands `~` for us before we see it, anyway. (I changed the type in `Cli` from `String` to `PathBuf`, to boot.) If we do need this sort of functionality in the future, https://docs.rs/shellexpand/latest/shellexpand/fn.tilde.html seems promising.
2025-04-25 14:20:21 -07:00
"dirs",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"env-flags",
"eventsource-stream",
"fs-err",
feat: record messages from user in ~/.codex/history.jsonl (#939) This is a large change to support a "history" feature like you would expect in a shell like Bash. History events are recorded in `$CODEX_HOME/history.jsonl`. Because it is a JSONL file, it is straightforward to append new entries (as opposed to the TypeScript file that uses `$CODEX_HOME/history.json`, so to be valid JSON, each new entry entails rewriting the entire file). Because it is possible for there to be multiple instances of Codex CLI writing to `history.jsonl` at once, we use advisory file locking when working with `history.jsonl` in `codex-rs/core/src/message_history.rs`. Because we believe history is a sufficiently useful feature, we enable it by default. Though to provide some safety, we set the file permissions of `history.jsonl` to be `o600` so that other users on the system cannot read the user's history. We do not yet support a default list of `SENSITIVE_PATTERNS` as the TypeScript CLI does: https://github.com/openai/codex/blob/3fdf9df1335ac9501e3fb0e61715359145711e8b/codex-cli/src/utils/storage/command-history.ts#L10-L17 We are going to take a more conservative approach to this list in the Rust CLI. For example, while `/\b[A-Za-z0-9-_]{20,}\b/` might exclude sensitive information like API tokens, it would also exclude valuable information such as references to Git commits. As noted in the updated documentation, users can opt-out of history by adding the following to `config.toml`: ```toml [history] persistence = "none" ``` Because `history.jsonl` could, in theory, be quite large, we take a[n arguably overly pedantic] approach in reading history entries into memory. Specifically, we start by telling the client the current number of entries in the history file (`history_entry_count`) as well as the inode (`history_log_id`) of `history.jsonl` (see the new fields on `SessionConfiguredEvent`). The client is responsible for keeping new entries in memory to create a "local history," but if the user hits up enough times to go "past" the end of local history, then the client should use the new `GetHistoryEntryRequest` in the protocol to fetch older entries. Specifically, it should pass the `history_log_id` it was given originally and work backwards from `history_entry_count`. (It should really fetch history in batches rather than one-at-a-time, but that is something we can improve upon in subsequent PRs.) The motivation behind this crazy scheme is that it is designed to defend against: * The `history.jsonl` being truncated during the session such that the index into the history is no longer consistent with what had been read up to that point. We do not yet have logic to enforce a `max_bytes` for `history.jsonl`, but once we do, we will aspire to implement it in a way that should result in a new inode for the file on most systems. * New items from concurrent Codex CLI sessions amending to the history. Because, in absence of truncation, `history.jsonl` is an append-only log, so long as the client reads backwards from `history_entry_count`, it should always get a consistent view of history. (That said, it will not be able to read _new_ commands from concurrent sessions, but perhaps we will introduce a `/` command to reload latest history or something down the road.) Admittedly, my testing of this feature thus far has been fairly light. I expect we will find bugs and introduce enhancements/fixes going forward.
2025-05-15 16:26:23 -07:00
"fs2",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"futures",
"landlock",
feat: introduce support for shell_environment_policy in config.toml (#1061) To date, when handling `shell` and `local_shell` tool calls, we were spawning new processes using the environment inherited from the Codex process itself. This means that the sensitive `OPENAI_API_KEY` that Codex needs to talk to OpenAI models was made available to everything run by `shell` and `local_shell`. While there are cases where that might be useful, it does not seem like a good default. This PR introduces a complex `shell_environment_policy` config option to control the `env` used with these tool calls. It is inevitably a bit complex so that it is possible to override individual components of the policy so without having to restate the entire thing. Details are in the updated `README.md` in this PR, but here is the relevant bit that explains the individual fields of `shell_environment_policy`: | Field | Type | Default | Description | | ------------------------- | -------------------------- | ------- | ----------------------------------------------------------------------------------------------------------------------------------------------- | | `inherit` | string | `core` | Starting template for the environment:<br>`core` (`HOME`, `PATH`, `USER`, …), `all` (clone full parent env), or `none` (start empty). | | `ignore_default_excludes` | boolean | `false` | When `false`, Codex removes any var whose **name** contains `KEY`, `SECRET`, or `TOKEN` (case-insensitive) before other rules run. | | `exclude` | array&lt;string&gt; | `[]` | Case-insensitive glob patterns to drop after the default filter.<br>Examples: `"AWS_*"`, `"AZURE_*"`. | | `set` | table&lt;string,string&gt; | `{}` | Explicit key/value overrides or additions – always win over inherited values. | | `include_only` | array&lt;string&gt; | `[]` | If non-empty, a whitelist of patterns; only variables that match _one_ pattern survive the final step. (Generally used with `inherit = "all"`.) | In particular, note that the default is `inherit = "core"`, so: * if you have extra env variables that you want to inherit from the parent process, use `inherit = "all"` and then specify `include_only` * if you have extra env variables where you want to hardcode the values, the default `inherit = "core"` will work fine, but then you need to specify `set` This configuration is not battle-tested, so we will probably still have to play with it a bit. `core/src/exec_env.rs` has the critical business logic as well as unit tests. Though if nothing else, previous to this change: ``` $ cargo run --bin codex -- debug seatbelt -- printenv OPENAI_API_KEY # ...prints OPENAI_API_KEY... ``` But after this change it does not print anything (as desired). One final thing to call out about this PR is that the `configure_command!` macro we use in `core/src/exec.rs` has to do some complex logic with respect to how it builds up the `env` for the process being spawned under Landlock/seccomp. Specifically, doing `cmd.env_clear()` followed by `cmd.envs(&$env_map)` (which is arguably the most intuitive way to do it) caused the Landlock unit tests to fail because the processes spawned by the unit tests started failing in unexpected ways! If we forgo `env_clear()` in favor of updating env vars one at a time, the tests still pass. The comment in the code talks about this a bit, and while I would like to investigate this more, I need to move on for the moment, but I do plan to come back to it to fully understand what is going on. For example, this suggests that we might not be able to spawn a C program that calls `env_clear()`, which would be...weird. We may still have to fiddle with our Landlock config if that is the case.
2025-05-22 09:51:19 -07:00
"maplit",
feat: support mcp_servers in config.toml (#829) This adds initial support for MCP servers in the style of Claude Desktop and Cursor. Note this PR is the bare minimum to get things working end to end: all configured MCP servers are launched every time Codex is run, there is no recovery for MCP servers that crash, etc. (Also, I took some shortcuts to change some fields of `Session` to be `pub(crate)`, which also means there are circular deps between `codex.rs` and `mcp_tool_call.rs`, but I will clean that up in a subsequent PR.) `codex-rs/README.md` is updated as part of this PR to explain how to use this feature. There is a bit of plumbing to route the new settings from `Config` to the business logic in `codex.rs`. The most significant chunks for new code are in `mcp_connection_manager.rs` (which defines the `McpConnectionManager` struct) and `mcp_tool_call.rs`, which is responsible for tool calls. This PR also introduces new `McpToolCallBegin` and `McpToolCallEnd` event types to the protocol, but does not add any handlers for them. (See https://github.com/openai/codex/pull/836 for initial usage.) To test, I added the following to my `~/.codex/config.toml`: ```toml # Local build of https://github.com/hideya/mcp-server-weather-js [mcp_servers.weather] command = "/Users/mbolin/code/mcp-server-weather-js/dist/index.js" args = [] ``` And then I ran the following: ``` codex-rs$ cargo run --bin codex exec 'what is the weather in san francisco' [2025-05-06T22:40:05] Task started: 1 [2025-05-06T22:40:18] Agent message: Here’s the latest National Weather Service forecast for San Francisco (downtown, near 37.77° N, 122.42° W): This Afternoon (Tue): • Sunny, high near 69 °F • West-southwest wind around 12 mph Tonight: • Partly cloudy, low around 52 °F • SW wind 7–10 mph ... ``` Note that Codex itself is not able to make network calls, so it would not normally be able to get live weather information like this. However, the weather MCP is [currently] not run under the Codex sandbox, so it is able to hit `api.weather.gov` and fetch current weather information. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/829). * #836 * __->__ #829
2025-05-06 15:47:59 -07:00
"mcp-types",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"mime_guess",
"openssl-sys",
"patch",
feat: bring back -s option to specify sandbox permissions (#739)
2025-04-29 18:42:52 -07:00
"path-absolutize",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"predicates",
feat: introduce --profile for Rust CLI (#921) This introduces a much-needed "profile" concept where users can specify a collection of options under one name and then pass that via `--profile` to the CLI. This PR introduces the `ConfigProfile` struct and makes it a field of `CargoToml`. It further updates `Config::load_from_base_config_with_overrides()` to respect `ConfigProfile`, overriding default values where appropriate. A detailed unit test is added at the end of `config.rs` to verify this behavior. Details on how to use this feature have also been added to `codex-rs/README.md`.
2025-05-13 16:52:52 -07:00
"pretty_assertions",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"rand 0.9.1",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"reqwest",
"seccompiler",
"serde",
"serde_json",
feat: make reasoning effort/summaries configurable (#1199) Previous to this PR, we always set `reasoning` when making a request using the Responses API: https://github.com/openai/codex/blob/d7245cbbc9d8ff5446da45e5951761103492476d/codex-rs/core/src/client.rs#L108-L111 Though if you tried to use the Rust CLI with `--model gpt-4.1`, this would fail with: ```shell "Unsupported parameter: 'reasoning.effort' is not supported with this model." ``` We take a cue from the TypeScript CLI, which does a check on the model name: https://github.com/openai/codex/blob/d7245cbbc9d8ff5446da45e5951761103492476d/codex-cli/src/utils/agent/agent-loop.ts#L786-L789 This PR does a similar check, though also adds support for the following config options: ``` model_reasoning_effort = "low" | "medium" | "high" | "none" model_reasoning_summary = "auto" | "concise" | "detailed" | "none" ``` This way, if you have a model whose name happens to start with `"o"` (or `"codex"`?), you can set these to `"none"` to explicitly disable reasoning, if necessary. (That said, it seems unlikely anyone would use the Responses API with non-OpenAI models, but we provide an escape hatch, anyway.) This PR also updates both the TUI and `codex exec` to show `reasoning effort` and `reasoning summaries` in the header.
2025-06-02 16:01:34 -07:00
"strum 0.27.1",
"strum_macros 0.27.1",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tempfile",
"thiserror 2.0.12",
feat: save session transcripts when using Rust CLI (#845) This adds support for saving transcripts when using the Rust CLI. Like the TypeScript CLI, it saves the transcript to `~/.codex/sessions`, though it uses JSONL for the file format (and `.jsonl` for the file extension) so that even if Codex crashes, what was written to the `.jsonl` file should generally still be valid JSONL content.
2025-05-07 13:49:15 -07:00
"time",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tokio",
"tokio-util",
"toml",
"tracing",
"tree-sitter",
"tree-sitter-bash",
feat: save session transcripts when using Rust CLI (#845) This adds support for saving transcripts when using the Rust CLI. Like the TypeScript CLI, it saves the transcript to `~/.codex/sessions`, though it uses JSONL for the file format (and `.jsonl` for the file extension) so that even if Codex crashes, what was written to the `.jsonl` file should generally still be valid JSONL content.
2025-05-07 13:49:15 -07:00
"uuid",
feat: introduce support for shell_environment_policy in config.toml (#1061) To date, when handling `shell` and `local_shell` tool calls, we were spawning new processes using the environment inherited from the Codex process itself. This means that the sensitive `OPENAI_API_KEY` that Codex needs to talk to OpenAI models was made available to everything run by `shell` and `local_shell`. While there are cases where that might be useful, it does not seem like a good default. This PR introduces a complex `shell_environment_policy` config option to control the `env` used with these tool calls. It is inevitably a bit complex so that it is possible to override individual components of the policy so without having to restate the entire thing. Details are in the updated `README.md` in this PR, but here is the relevant bit that explains the individual fields of `shell_environment_policy`: | Field | Type | Default | Description | | ------------------------- | -------------------------- | ------- | ----------------------------------------------------------------------------------------------------------------------------------------------- | | `inherit` | string | `core` | Starting template for the environment:<br>`core` (`HOME`, `PATH`, `USER`, …), `all` (clone full parent env), or `none` (start empty). | | `ignore_default_excludes` | boolean | `false` | When `false`, Codex removes any var whose **name** contains `KEY`, `SECRET`, or `TOKEN` (case-insensitive) before other rules run. | | `exclude` | array&lt;string&gt; | `[]` | Case-insensitive glob patterns to drop after the default filter.<br>Examples: `"AWS_*"`, `"AZURE_*"`. | | `set` | table&lt;string,string&gt; | `{}` | Explicit key/value overrides or additions – always win over inherited values. | | `include_only` | array&lt;string&gt; | `[]` | If non-empty, a whitelist of patterns; only variables that match _one_ pattern survive the final step. (Generally used with `inherit = "all"`.) | In particular, note that the default is `inherit = "core"`, so: * if you have extra env variables that you want to inherit from the parent process, use `inherit = "all"` and then specify `include_only` * if you have extra env variables where you want to hardcode the values, the default `inherit = "core"` will work fine, but then you need to specify `set` This configuration is not battle-tested, so we will probably still have to play with it a bit. `core/src/exec_env.rs` has the critical business logic as well as unit tests. Though if nothing else, previous to this change: ``` $ cargo run --bin codex -- debug seatbelt -- printenv OPENAI_API_KEY # ...prints OPENAI_API_KEY... ``` But after this change it does not print anything (as desired). One final thing to call out about this PR is that the `configure_command!` macro we use in `core/src/exec.rs` has to do some complex logic with respect to how it builds up the `env` for the process being spawned under Landlock/seccomp. Specifically, doing `cmd.env_clear()` followed by `cmd.envs(&$env_map)` (which is arguably the most intuitive way to do it) caused the Landlock unit tests to fail because the processes spawned by the unit tests started failing in unexpected ways! If we forgo `env_clear()` in favor of updating env vars one at a time, the tests still pass. The comment in the code talks about this a bit, and while I would like to investigate this more, I need to move on for the moment, but I do plan to come back to it to fully understand what is going on. For example, this suggests that we might not be able to spawn a C program that calls `env_clear()`, which would be...weird. We may still have to fiddle with our Landlock config if that is the case.
2025-05-22 09:51:19 -07:00
"wildmatch",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"wiremock",
]
[[package]]
name = "codex-exec"
chore: script to create a Rust release (#759) For now, keep things simple such that we never update the `version` in the `Cargo.toml` for the workspace root on the `main` branch. Instead, create a new branch for a release, push one commit that updates the `version`, and then tag that branch to kick off a release. To test, I ran this script and created this release job: https://github.com/openai/codex/actions/runs/14762580641
2025-04-30 12:39:03 -07:00
version = "0.0.0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
dependencies = [
"anyhow",
feat: improve output of exec subcommand (#719)
2025-04-29 09:59:35 -07:00
"chrono",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"clap",
chore: introduce codex-common crate (#843) I started this PR because I wanted to share the `format_duration()` utility function in `codex-rs/exec/src/event_processor.rs` with the TUI. The question was: where to put it? `core` should have as few dependencies as possible, so moving it there would introduce a dependency on `chrono`, which seemed undesirable. `core` already had this `cli` feature to deal with a similar situation around sharing common utility functions, so I decided to: * make `core` feature-free * introduce `common` * `common` can have as many "special interest" features as it needs, each of which can declare their own deps * the first two features of common are `cli` and `elapsed` In practice, this meant updating a number of `Cargo.toml` files, replacing this line: ```toml codex-core = { path = "../core", features = ["cli"] } ``` with these: ```toml codex-core = { path = "../core" } codex-common = { path = "../common", features = ["cli"] } ``` Moving `format_duration()` into its own file gave it some "breathing room" to add a unit test, so I had Codex generate some tests and new support for durations over 1 minute.
2025-05-06 17:38:56 -07:00
"codex-common",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"codex-core",
fix: overhaul how we spawn commands under seccomp/landlock on Linux (#1086) Historically, we spawned the Seatbelt and Landlock sandboxes in substantially different ways: For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy specified as an arg followed by the original command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec.rs#L147-L219 For **Landlock/Seccomp**, we would do `tokio::runtime::Builder::new_current_thread()`, _invoke Landlock/Seccomp APIs to modify the permissions of that new thread_, and then spawn the command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec_linux.rs#L28-L49 While it is neat that Landlock/Seccomp supports applying a policy to only one thread without having to apply it to the entire process, it requires us to maintain two different codepaths and is a bit harder to reason about. The tipping point was https://github.com/openai/codex/pull/1061, in which we had to start building up the `env` in an unexpected way for the existing Landlock/Seccomp approach to continue to work. This PR overhauls things so that we do similar things for Mac and Linux. It turned out that we were already building our own "helper binary" comparable to Mac's `sandbox-exec` as part of the `cli` crate: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/cli/Cargo.toml#L10-L12 We originally created this to build a small binary to include with the Node.js version of the Codex CLI to provide support for Linux sandboxing. Though the sticky bit is that, at this point, we still want to deploy the Rust version of Codex as a single, standalone binary rather than a CLI and a supporting sandboxing binary. To satisfy this goal, we use "the arg0 trick," in which we: * use `std::env::current_exe()` to get the path to the CLI that is currently running * use the CLI as the `program` for the `Command` * set `"codex-linux-sandbox"` as arg0 for the `Command` A CLI that supports sandboxing should check arg0 at the start of the program. If it is `"codex-linux-sandbox"`, it must invoke `codex_linux_sandbox::run_main()`, which runs the CLI as if it were `codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn the original command, so do _replace_ the process rather than spawn a subprocess. Incidentally, we do this before starting the Tokio runtime, so the process should only have one thread when `execvp(3)` is called. Because the `core` crate that needs to spawn the Linux sandboxing is not a CLI in its own right, this means that every CLI that includes `core` and relies on this behavior has to (1) implement it and (2) provide the path to the sandboxing executable. While the path is almost always `std::env::current_exe()`, we needed to make this configurable for integration tests, so `Config` now has a `codex_linux_sandbox_exe: Option<PathBuf>` property to facilitate threading this through, introduced in https://github.com/openai/codex/pull/1089. This common pattern is now captured in `codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs` functions that should use it have been updated as part of this PR. The `codex-linux-sandbox` crate added to the Cargo workspace as part of this PR now has the bulk of the Landlock/Seccomp logic, which makes `core` a bit simpler. Indeed, `core/src/exec_linux.rs` and `core/src/landlock.rs` were removed/ported as part of this PR. I also moved the unit tests for this code into an integration test, `linux-sandbox/tests/landlock.rs`, in which I use `env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for `codex_linux_sandbox_exe` since `std::env::current_exe()` is not appropriate in that case.
2025-05-23 11:37:07 -07:00
"codex-linux-sandbox",
feat: show MCP tool calls in `codex exec` subcommand (#841) This is analogous to the change for the TUI in https://github.com/openai/codex/pull/836, but for `codex exec`. To test, I ran: ``` cargo run --bin codex-exec -- 'what is the weather in wellesley ma tomorrow' ``` and saw: ![image](https://github.com/user-attachments/assets/5714e07f-88c7-4dd9-aa0d-be54c1670533)
2025-05-06 16:52:43 -07:00
"mcp-types",
feat: improve output of exec subcommand (#719)
2025-04-29 09:59:35 -07:00
"owo-colors 4.2.0",
feat: show MCP tool calls in `codex exec` subcommand (#841) This is analogous to the change for the TUI in https://github.com/openai/codex/pull/836, but for `codex exec`. To test, I ran: ``` cargo run --bin codex-exec -- 'what is the weather in wellesley ma tomorrow' ``` and saw: ![image](https://github.com/user-attachments/assets/5714e07f-88c7-4dd9-aa0d-be54c1670533)
2025-05-06 16:52:43 -07:00
"serde_json",
feat: improve output of exec subcommand (#719)
2025-04-29 09:59:35 -07:00
"shlex",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tokio",
"tracing",
"tracing-subscriber",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "codex-execpolicy"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
dependencies = [
"allocative",
"anyhow",
"clap",
"derive_more",
"env_logger",
"log",
"multimap",
"path-absolutize",
chore: replace regex with regex-lite, where appropriate (#1200) As explained on https://crates.io/crates/regex-lite, `regex-lite` is a lighter alternative to `regex` and seems to be sufficient for our purposes.
2025-06-02 17:11:45 -07:00
"regex-lite",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"serde",
"serde_json",
"serde_with",
"starlark",
"tempfile",
]
fix: overhaul how we spawn commands under seccomp/landlock on Linux (#1086) Historically, we spawned the Seatbelt and Landlock sandboxes in substantially different ways: For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy specified as an arg followed by the original command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec.rs#L147-L219 For **Landlock/Seccomp**, we would do `tokio::runtime::Builder::new_current_thread()`, _invoke Landlock/Seccomp APIs to modify the permissions of that new thread_, and then spawn the command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec_linux.rs#L28-L49 While it is neat that Landlock/Seccomp supports applying a policy to only one thread without having to apply it to the entire process, it requires us to maintain two different codepaths and is a bit harder to reason about. The tipping point was https://github.com/openai/codex/pull/1061, in which we had to start building up the `env` in an unexpected way for the existing Landlock/Seccomp approach to continue to work. This PR overhauls things so that we do similar things for Mac and Linux. It turned out that we were already building our own "helper binary" comparable to Mac's `sandbox-exec` as part of the `cli` crate: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/cli/Cargo.toml#L10-L12 We originally created this to build a small binary to include with the Node.js version of the Codex CLI to provide support for Linux sandboxing. Though the sticky bit is that, at this point, we still want to deploy the Rust version of Codex as a single, standalone binary rather than a CLI and a supporting sandboxing binary. To satisfy this goal, we use "the arg0 trick," in which we: * use `std::env::current_exe()` to get the path to the CLI that is currently running * use the CLI as the `program` for the `Command` * set `"codex-linux-sandbox"` as arg0 for the `Command` A CLI that supports sandboxing should check arg0 at the start of the program. If it is `"codex-linux-sandbox"`, it must invoke `codex_linux_sandbox::run_main()`, which runs the CLI as if it were `codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn the original command, so do _replace_ the process rather than spawn a subprocess. Incidentally, we do this before starting the Tokio runtime, so the process should only have one thread when `execvp(3)` is called. Because the `core` crate that needs to spawn the Linux sandboxing is not a CLI in its own right, this means that every CLI that includes `core` and relies on this behavior has to (1) implement it and (2) provide the path to the sandboxing executable. While the path is almost always `std::env::current_exe()`, we needed to make this configurable for integration tests, so `Config` now has a `codex_linux_sandbox_exe: Option<PathBuf>` property to facilitate threading this through, introduced in https://github.com/openai/codex/pull/1089. This common pattern is now captured in `codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs` functions that should use it have been updated as part of this PR. The `codex-linux-sandbox` crate added to the Cargo workspace as part of this PR now has the bulk of the Landlock/Seccomp logic, which makes `core` a bit simpler. Indeed, `core/src/exec_linux.rs` and `core/src/landlock.rs` were removed/ported as part of this PR. I also moved the unit tests for this code into an integration test, `linux-sandbox/tests/landlock.rs`, in which I use `env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for `codex_linux_sandbox_exe` since `std::env::current_exe()` is not appropriate in that case.
2025-05-23 11:37:07 -07:00
[[package]]
name = "codex-linux-sandbox"
version = "0.0.0"
dependencies = [
"anyhow",
"clap",
"codex-common",
"codex-core",
"landlock",
"libc",
"seccompiler",
"tempfile",
"tokio",
]
feat: initial McpClient for Rust (#822) This PR introduces an initial `McpClient` that we will use to give Codex itself programmatic access to foreign MCPs. This does not wire it up in Codex itself yet, but the new `mcp-client` crate includes a `main.rs` for basic testing for now. Manually tested by sending a `tools/list` request to Codex's own MCP server: ``` codex-rs$ cargo build codex-rs$ cargo run --bin codex-mcp-client ./target/debug/codex-mcp-server { "tools": [ { "description": "Run a Codex session. Accepts configuration parameters matching the Codex Config struct.", "inputSchema": { "properties": { "approval-policy": { "description": "Execution approval policy expressed as the kebab-case variant name (`unless-allow-listed`, `auto-edit`, `on-failure`, `never`).", "enum": [ "auto-edit", "unless-allow-listed", "on-failure", "never" ], "type": "string" }, "cwd": { "description": "Working directory for the session. If relative, it is resolved against the server process's current working directory.", "type": "string" }, "disable-response-storage": { "description": "Disable server-side response storage.", "type": "boolean" }, "model": { "description": "Optional override for the model name (e.g. \"o3\", \"o4-mini\")", "type": "string" }, "prompt": { "description": "The *initial user prompt* to start the Codex conversation.", "type": "string" }, "sandbox-permissions": { "description": "Sandbox permissions using the same string values accepted by the CLI (e.g. \"disk-write-cwd\", \"network-full-access\").", "items": { "enum": [ "disk-full-read-access", "disk-write-cwd", "disk-write-platform-user-temp-folder", "disk-write-platform-global-temp-folder", "disk-full-write-access", "network-full-access" ], "type": "string" }, "type": "array" } }, "required": [ "prompt" ], "type": "object" }, "name": "codex" } ] } ```
2025-05-05 12:52:55 -07:00
[[package]]
name = "codex-mcp-client"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: initial McpClient for Rust (#822) This PR introduces an initial `McpClient` that we will use to give Codex itself programmatic access to foreign MCPs. This does not wire it up in Codex itself yet, but the new `mcp-client` crate includes a `main.rs` for basic testing for now. Manually tested by sending a `tools/list` request to Codex's own MCP server: ``` codex-rs$ cargo build codex-rs$ cargo run --bin codex-mcp-client ./target/debug/codex-mcp-server { "tools": [ { "description": "Run a Codex session. Accepts configuration parameters matching the Codex Config struct.", "inputSchema": { "properties": { "approval-policy": { "description": "Execution approval policy expressed as the kebab-case variant name (`unless-allow-listed`, `auto-edit`, `on-failure`, `never`).", "enum": [ "auto-edit", "unless-allow-listed", "on-failure", "never" ], "type": "string" }, "cwd": { "description": "Working directory for the session. If relative, it is resolved against the server process's current working directory.", "type": "string" }, "disable-response-storage": { "description": "Disable server-side response storage.", "type": "boolean" }, "model": { "description": "Optional override for the model name (e.g. \"o3\", \"o4-mini\")", "type": "string" }, "prompt": { "description": "The *initial user prompt* to start the Codex conversation.", "type": "string" }, "sandbox-permissions": { "description": "Sandbox permissions using the same string values accepted by the CLI (e.g. \"disk-write-cwd\", \"network-full-access\").", "items": { "enum": [ "disk-full-read-access", "disk-write-cwd", "disk-write-platform-user-temp-folder", "disk-write-platform-global-temp-folder", "disk-full-write-access", "network-full-access" ], "type": "string" }, "type": "array" } }, "required": [ "prompt" ], "type": "object" }, "name": "codex" } ] } ```
2025-05-05 12:52:55 -07:00
dependencies = [
"anyhow",
"mcp-types",
"pretty_assertions",
"serde",
"serde_json",
"tokio",
"tracing",
"tracing-subscriber",
]
feat: introduce mcp-server crate (#792) This introduces the `mcp-server` crate, which contains a barebones MCP server that provides an `echo` tool that echoes the user's request back to them. To test it out, I launched [modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector) like so: ``` mcp-server$ npx @modelcontextprotocol/inspector cargo run -- ``` and opened up `http://127.0.0.1:6274` in my browser: ![image](https://github.com/user-attachments/assets/83fc55d4-25c2-4497-80cd-e9702283ff93) I also had to make a small fix to `mcp-types`, adding `#[serde(untagged)]` to a number of `enum`s.
2025-05-02 17:25:58 -07:00
[[package]]
name = "codex-mcp-server"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: introduce mcp-server crate (#792) This introduces the `mcp-server` crate, which contains a barebones MCP server that provides an `echo` tool that echoes the user's request back to them. To test it out, I launched [modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector) like so: ``` mcp-server$ npx @modelcontextprotocol/inspector cargo run -- ``` and opened up `http://127.0.0.1:6274` in my browser: ![image](https://github.com/user-attachments/assets/83fc55d4-25c2-4497-80cd-e9702283ff93) I also had to make a small fix to `mcp-types`, adding `#[serde(untagged)]` to a number of `enum`s.
2025-05-02 17:25:58 -07:00
dependencies = [
fix: overhaul how we spawn commands under seccomp/landlock on Linux (#1086) Historically, we spawned the Seatbelt and Landlock sandboxes in substantially different ways: For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy specified as an arg followed by the original command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec.rs#L147-L219 For **Landlock/Seccomp**, we would do `tokio::runtime::Builder::new_current_thread()`, _invoke Landlock/Seccomp APIs to modify the permissions of that new thread_, and then spawn the command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec_linux.rs#L28-L49 While it is neat that Landlock/Seccomp supports applying a policy to only one thread without having to apply it to the entire process, it requires us to maintain two different codepaths and is a bit harder to reason about. The tipping point was https://github.com/openai/codex/pull/1061, in which we had to start building up the `env` in an unexpected way for the existing Landlock/Seccomp approach to continue to work. This PR overhauls things so that we do similar things for Mac and Linux. It turned out that we were already building our own "helper binary" comparable to Mac's `sandbox-exec` as part of the `cli` crate: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/cli/Cargo.toml#L10-L12 We originally created this to build a small binary to include with the Node.js version of the Codex CLI to provide support for Linux sandboxing. Though the sticky bit is that, at this point, we still want to deploy the Rust version of Codex as a single, standalone binary rather than a CLI and a supporting sandboxing binary. To satisfy this goal, we use "the arg0 trick," in which we: * use `std::env::current_exe()` to get the path to the CLI that is currently running * use the CLI as the `program` for the `Command` * set `"codex-linux-sandbox"` as arg0 for the `Command` A CLI that supports sandboxing should check arg0 at the start of the program. If it is `"codex-linux-sandbox"`, it must invoke `codex_linux_sandbox::run_main()`, which runs the CLI as if it were `codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn the original command, so do _replace_ the process rather than spawn a subprocess. Incidentally, we do this before starting the Tokio runtime, so the process should only have one thread when `execvp(3)` is called. Because the `core` crate that needs to spawn the Linux sandboxing is not a CLI in its own right, this means that every CLI that includes `core` and relies on this behavior has to (1) implement it and (2) provide the path to the sandboxing executable. While the path is almost always `std::env::current_exe()`, we needed to make this configurable for integration tests, so `Config` now has a `codex_linux_sandbox_exe: Option<PathBuf>` property to facilitate threading this through, introduced in https://github.com/openai/codex/pull/1089. This common pattern is now captured in `codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs` functions that should use it have been updated as part of this PR. The `codex-linux-sandbox` crate added to the Cargo workspace as part of this PR now has the bulk of the Landlock/Seccomp logic, which makes `core` a bit simpler. Indeed, `core/src/exec_linux.rs` and `core/src/landlock.rs` were removed/ported as part of this PR. I also moved the unit tests for this code into an integration test, `linux-sandbox/tests/landlock.rs`, in which I use `env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for `codex_linux_sandbox_exe` since `std::env::current_exe()` is not appropriate in that case.
2025-05-23 11:37:07 -07:00
"anyhow",
feat: introduce mcp-server crate (#792) This introduces the `mcp-server` crate, which contains a barebones MCP server that provides an `echo` tool that echoes the user's request back to them. To test it out, I launched [modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector) like so: ``` mcp-server$ npx @modelcontextprotocol/inspector cargo run -- ``` and opened up `http://127.0.0.1:6274` in my browser: ![image](https://github.com/user-attachments/assets/83fc55d4-25c2-4497-80cd-e9702283ff93) I also had to make a small fix to `mcp-types`, adding `#[serde(untagged)]` to a number of `enum`s.
2025-05-02 17:25:58 -07:00
"codex-core",
fix: overhaul how we spawn commands under seccomp/landlock on Linux (#1086) Historically, we spawned the Seatbelt and Landlock sandboxes in substantially different ways: For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy specified as an arg followed by the original command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec.rs#L147-L219 For **Landlock/Seccomp**, we would do `tokio::runtime::Builder::new_current_thread()`, _invoke Landlock/Seccomp APIs to modify the permissions of that new thread_, and then spawn the command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec_linux.rs#L28-L49 While it is neat that Landlock/Seccomp supports applying a policy to only one thread without having to apply it to the entire process, it requires us to maintain two different codepaths and is a bit harder to reason about. The tipping point was https://github.com/openai/codex/pull/1061, in which we had to start building up the `env` in an unexpected way for the existing Landlock/Seccomp approach to continue to work. This PR overhauls things so that we do similar things for Mac and Linux. It turned out that we were already building our own "helper binary" comparable to Mac's `sandbox-exec` as part of the `cli` crate: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/cli/Cargo.toml#L10-L12 We originally created this to build a small binary to include with the Node.js version of the Codex CLI to provide support for Linux sandboxing. Though the sticky bit is that, at this point, we still want to deploy the Rust version of Codex as a single, standalone binary rather than a CLI and a supporting sandboxing binary. To satisfy this goal, we use "the arg0 trick," in which we: * use `std::env::current_exe()` to get the path to the CLI that is currently running * use the CLI as the `program` for the `Command` * set `"codex-linux-sandbox"` as arg0 for the `Command` A CLI that supports sandboxing should check arg0 at the start of the program. If it is `"codex-linux-sandbox"`, it must invoke `codex_linux_sandbox::run_main()`, which runs the CLI as if it were `codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn the original command, so do _replace_ the process rather than spawn a subprocess. Incidentally, we do this before starting the Tokio runtime, so the process should only have one thread when `execvp(3)` is called. Because the `core` crate that needs to spawn the Linux sandboxing is not a CLI in its own right, this means that every CLI that includes `core` and relies on this behavior has to (1) implement it and (2) provide the path to the sandboxing executable. While the path is almost always `std::env::current_exe()`, we needed to make this configurable for integration tests, so `Config` now has a `codex_linux_sandbox_exe: Option<PathBuf>` property to facilitate threading this through, introduced in https://github.com/openai/codex/pull/1089. This common pattern is now captured in `codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs` functions that should use it have been updated as part of this PR. The `codex-linux-sandbox` crate added to the Cargo workspace as part of this PR now has the bulk of the Landlock/Seccomp logic, which makes `core` a bit simpler. Indeed, `core/src/exec_linux.rs` and `core/src/landlock.rs` were removed/ported as part of this PR. I also moved the unit tests for this code into an integration test, `linux-sandbox/tests/landlock.rs`, in which I use `env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for `codex_linux_sandbox_exe` since `std::env::current_exe()` is not appropriate in that case.
2025-05-23 11:37:07 -07:00
"codex-linux-sandbox",
feat: introduce mcp-server crate (#792) This introduces the `mcp-server` crate, which contains a barebones MCP server that provides an `echo` tool that echoes the user's request back to them. To test it out, I launched [modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector) like so: ``` mcp-server$ npx @modelcontextprotocol/inspector cargo run -- ``` and opened up `http://127.0.0.1:6274` in my browser: ![image](https://github.com/user-attachments/assets/83fc55d4-25c2-4497-80cd-e9702283ff93) I also had to make a small fix to `mcp-types`, adding `#[serde(untagged)]` to a number of `enum`s.
2025-05-02 17:25:58 -07:00
"mcp-types",
feat: make Codex available as a tool when running it as an MCP server (#811) This PR replaces the placeholder `"echo"` tool call in the MCP server with a `"codex"` tool that calls Codex. Events such as `ExecApprovalRequest` and `ApplyPatchApprovalRequest` are not handled properly yet, but I have `approval_policy = "never"` set in my `~/.codex/config.toml` such that those codepaths are not exercised. The schema for this MPC tool is defined by a new `CodexToolCallParam` struct introduced in this PR. It is fairly similar to `ConfigOverrides`, as the param is used to help create the `Config` used to start the Codex session, though it also includes the `prompt` used to kick off the session. This PR also introduces the use of the third-party `schemars` crate to generate the JSON schema, which is verified in the `verify_codex_tool_json_schema()` unit test. Events that are dispatched during the Codex session are sent back to the MCP client as MCP notifications. This gives the client a way to monitor progress as the tool call itself may take minutes to complete depending on the complexity of the task requested by the user. In the video below, I launched the server via: ```shell mcp-server$ RUST_LOG=debug npx @modelcontextprotocol/inspector cargo run -- ``` In the video, you can see the flow of: * requesting the list of tools * choosing the **codex** tool * entering a value for **prompt** and then making the tool call Note that I left the other fields blank because when unspecified, the values in my `~/.codex/config.toml` were used: https://github.com/user-attachments/assets/1975058c-b004-43ef-8c8d-800a953b8192 Note that while using the inspector, I did run into https://github.com/modelcontextprotocol/inspector/issues/293, though the tip about ensuring I had only one instance of the **MCP Inspector** tab open in my browser seemed to fix things.
2025-05-05 07:16:19 -07:00
"pretty_assertions",
"schemars",
feat: introduce mcp-server crate (#792) This introduces the `mcp-server` crate, which contains a barebones MCP server that provides an `echo` tool that echoes the user's request back to them. To test it out, I launched [modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector) like so: ``` mcp-server$ npx @modelcontextprotocol/inspector cargo run -- ``` and opened up `http://127.0.0.1:6274` in my browser: ![image](https://github.com/user-attachments/assets/83fc55d4-25c2-4497-80cd-e9702283ff93) I also had to make a small fix to `mcp-types`, adding `#[serde(untagged)]` to a number of `enum`s.
2025-05-02 17:25:58 -07:00
"serde",
"serde_json",
"tokio",
feat: add support for -c/--config to override individual config items (#1137) This PR introduces support for `-c`/`--config` so users can override individual config values on the command line using `--config name=value`. Example: ``` codex --config model=o4-mini ``` Making it possible to set arbitrary config values on the command line results in a more flexible configuration scheme and makes it easier to provide single-line examples that can be copy-pasted from documentation. Effectively, it means there are four levels of configuration for some values: - Default value (e.g., `model` currently defaults to `o4-mini`) - Value in `config.toml` (e.g., user could override the default to be `model = "o3"` in their `config.toml`) - Specifying `-c` or `--config` to override `model` (e.g., user can include `-c model=o3` in their list of args to Codex) - If available, a config-specific flag can be used, which takes precedence over `-c` (e.g., user can specify `--model o3` in their list of args to Codex) Now that it is possible to specify anything that could be configured in `config.toml` on the command line using `-c`, we do not need to have a custom flag for every possible config option (which can clutter the output of `--help`). To that end, as part of this PR, we drop support for the `--disable-response-storage` flag, as users can now specify `-c disable_response_storage=true` to get the equivalent functionality. Under the hood, this works by loading the `config.toml` into a `toml::Value`. Then for each `key=value`, we create a small synthetic TOML file with `value` so that we can run the TOML parser to get the equivalent `toml::Value`. We then parse `key` to determine the point in the original `toml::Value` to do the insert/replace. Once all of the overrides from `-c` args have been applied, the `toml::Value` is deserialized into a `ConfigToml` and then the `ConfigOverrides` are applied, as before.
2025-05-27 23:11:44 -07:00
"toml",
feat: introduce mcp-server crate (#792) This introduces the `mcp-server` crate, which contains a barebones MCP server that provides an `echo` tool that echoes the user's request back to them. To test it out, I launched [modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector) like so: ``` mcp-server$ npx @modelcontextprotocol/inspector cargo run -- ``` and opened up `http://127.0.0.1:6274` in my browser: ![image](https://github.com/user-attachments/assets/83fc55d4-25c2-4497-80cd-e9702283ff93) I also had to make a small fix to `mcp-types`, adding `#[serde(untagged)]` to a number of `enum`s.
2025-05-02 17:25:58 -07:00
"tracing",
"tracing-subscriber",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "codex-tui"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
dependencies = [
"anyhow",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"base64 0.22.1",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"clap",
"codex-ansi-escape",
chore: introduce codex-common crate (#843) I started this PR because I wanted to share the `format_duration()` utility function in `codex-rs/exec/src/event_processor.rs` with the TUI. The question was: where to put it? `core` should have as few dependencies as possible, so moving it there would introduce a dependency on `chrono`, which seemed undesirable. `core` already had this `cli` feature to deal with a similar situation around sharing common utility functions, so I decided to: * make `core` feature-free * introduce `common` * `common` can have as many "special interest" features as it needs, each of which can declare their own deps * the first two features of common are `cli` and `elapsed` In practice, this meant updating a number of `Cargo.toml` files, replacing this line: ```toml codex-core = { path = "../core", features = ["cli"] } ``` with these: ```toml codex-core = { path = "../core" } codex-common = { path = "../common", features = ["cli"] } ``` Moving `format_duration()` into its own file gave it some "breathing room" to add a unit test, so I had Codex generate some tests and new support for durations over 1 minute.
2025-05-06 17:38:56 -07:00
"codex-common",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"codex-core",
fix: overhaul how we spawn commands under seccomp/landlock on Linux (#1086) Historically, we spawned the Seatbelt and Landlock sandboxes in substantially different ways: For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy specified as an arg followed by the original command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec.rs#L147-L219 For **Landlock/Seccomp**, we would do `tokio::runtime::Builder::new_current_thread()`, _invoke Landlock/Seccomp APIs to modify the permissions of that new thread_, and then spawn the command: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/core/src/exec_linux.rs#L28-L49 While it is neat that Landlock/Seccomp supports applying a policy to only one thread without having to apply it to the entire process, it requires us to maintain two different codepaths and is a bit harder to reason about. The tipping point was https://github.com/openai/codex/pull/1061, in which we had to start building up the `env` in an unexpected way for the existing Landlock/Seccomp approach to continue to work. This PR overhauls things so that we do similar things for Mac and Linux. It turned out that we were already building our own "helper binary" comparable to Mac's `sandbox-exec` as part of the `cli` crate: https://github.com/openai/codex/blob/d1de7bb383552e8fadd94be79d65d188e00fd562/codex-rs/cli/Cargo.toml#L10-L12 We originally created this to build a small binary to include with the Node.js version of the Codex CLI to provide support for Linux sandboxing. Though the sticky bit is that, at this point, we still want to deploy the Rust version of Codex as a single, standalone binary rather than a CLI and a supporting sandboxing binary. To satisfy this goal, we use "the arg0 trick," in which we: * use `std::env::current_exe()` to get the path to the CLI that is currently running * use the CLI as the `program` for the `Command` * set `"codex-linux-sandbox"` as arg0 for the `Command` A CLI that supports sandboxing should check arg0 at the start of the program. If it is `"codex-linux-sandbox"`, it must invoke `codex_linux_sandbox::run_main()`, which runs the CLI as if it were `codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn the original command, so do _replace_ the process rather than spawn a subprocess. Incidentally, we do this before starting the Tokio runtime, so the process should only have one thread when `execvp(3)` is called. Because the `core` crate that needs to spawn the Linux sandboxing is not a CLI in its own right, this means that every CLI that includes `core` and relies on this behavior has to (1) implement it and (2) provide the path to the sandboxing executable. While the path is almost always `std::env::current_exe()`, we needed to make this configurable for integration tests, so `Config` now has a `codex_linux_sandbox_exe: Option<PathBuf>` property to facilitate threading this through, introduced in https://github.com/openai/codex/pull/1089. This common pattern is now captured in `codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs` functions that should use it have been updated as part of this PR. The `codex-linux-sandbox` crate added to the Cargo workspace as part of this PR now has the bulk of the Landlock/Seccomp logic, which makes `core` a bit simpler. Indeed, `core/src/exec_linux.rs` and `core/src/landlock.rs` were removed/ported as part of this PR. I also moved the unit tests for this code into an integration test, `linux-sandbox/tests/landlock.rs`, in which I use `env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for `codex_linux_sandbox_exe` since `std::env::current_exe()` is not appropriate in that case.
2025-05-23 11:37:07 -07:00
"codex-linux-sandbox",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"color-eyre",
"crossterm",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"image",
feat: add support for file_opener option in Rust, similiar to #911 (#957) This ports the enhancement introduced in https://github.com/openai/codex/pull/911 (and the fixes in https://github.com/openai/codex/pull/919) for the TypeScript CLI to the Rust one.
2025-05-16 11:33:08 -07:00
"lazy_static",
feat: show MCP tool calls in TUI (#836) Adds logic for the `McpToolCallBegin` and `McpToolCallEnd` events in `codex-rs/tui/src/chatwidget.rs` so they get entries in the conversation history in the TUI. Building on top of https://github.com/openai/codex/pull/829, here is the result of running: ``` cargo run --bin codex -- 'what is the weather in san francisco tomorrow' ``` ![image](https://github.com/user-attachments/assets/db4a79bb-4988-46cb-acb2-446d5ba9e058)
2025-05-06 16:12:15 -07:00
"mcp-types",
feat: add support for file_opener option in Rust, similiar to #911 (#957) This ports the enhancement introduced in https://github.com/openai/codex/pull/911 (and the fixes in https://github.com/openai/codex/pull/919) for the TypeScript CLI to the Rust one.
2025-05-16 11:33:08 -07:00
"path-clean",
"pretty_assertions",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"ratatui",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"ratatui-image",
chore: replace regex with regex-lite, where appropriate (#1200) As explained on https://crates.io/crates/regex-lite, `regex-lite` is a lighter alternative to `regex` and seems to be sufficient for our purposes.
2025-06-02 17:11:45 -07:00
"regex-lite",
feat: show MCP tool calls in TUI (#836) Adds logic for the `McpToolCallBegin` and `McpToolCallEnd` events in `codex-rs/tui/src/chatwidget.rs` so they get entries in the conversation history in the TUI. Building on top of https://github.com/openai/codex/pull/829, here is the result of running: ``` cargo run --bin codex -- 'what is the weather in san francisco tomorrow' ``` ![image](https://github.com/user-attachments/assets/db4a79bb-4988-46cb-acb2-446d5ba9e058)
2025-05-06 16:12:15 -07:00
"serde_json",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"shlex",
feat: add support for commands in the Rust TUI (#935) Introduces support for slash commands like in the TypeScript CLI. We do not support the full set of commands yet, but the core abstraction is there now. In particular, we have a `SlashCommand` enum and due to thoughtful use of the [strum](https://crates.io/crates/strum) crate, it requires minimal boilerplate to add a new command to the list. The key new piece of UI is `CommandPopup`, though the keyboard events are still handled by `ChatComposer`. The behavior is roughly as follows: * if the first character in the composer is `/`, the command popup is displayed (if you really want to send a message to Codex that starts with a `/`, simply put a space before the `/`) * while the popup is displayed, up/down can be used to change the selection of the popup * if there is a selection, hitting tab completes the command, but does not send it * if there is a selection, hitting enter sends the command * if the prefix of the composer matches a command, the command will be visible in the popup so the user can see the description (commands could take arguments, so additional text may appear after the command name itself) https://github.com/user-attachments/assets/39c3e6ee-eeb7-4ef7-a911-466d8184975f Incidentally, Codex wrote almost all the code for this PR!
2025-05-14 12:55:49 -07:00
"strum 0.27.1",
"strum_macros 0.27.1",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tokio",
"tracing",
"tracing-appender",
"tracing-subscriber",
"tui-input",
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
"tui-markdown",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tui-textarea",
codex-rs: make tool calls prettier (#1211) This PR overhauls how active tool calls and completed tool calls are displayed: 1. More use of colour to indicate success/failure and distinguish between components like tool name+arguments 2. Previously, the entire `CallToolResult` was serialized to JSON and pretty-printed. Now, we extract each individual `CallToolResultContent` and print those 1. The previous solution was wasting space by unnecessarily showing details of the `CallToolResult` struct to users, without formatting the actual tool call results nicely 2. We're now able to show users more information from tool results in less space, with nicer formatting when tools return JSON results ### Before: <img width="1251" alt="Screenshot 2025-06-03 at 11 24 26" src="https://github.com/user-attachments/assets/5a58f222-219c-4c53-ace7-d887194e30cf" /> ### After: <img width="1265" alt="image" src="https://github.com/user-attachments/assets/99fe54d0-9ebe-406a-855b-7aa529b91274" /> ## Future Work 1. Integrate image tool result handling better. We should be able to display images even if they're not the first `CallToolResultContent` 2. Users should have some way to view the full version of truncated tool results 3. It would be nice to add some left padding for tool results, make it more clear that they are results. This is doable, just a little fiddly due to the way `first_visible_line` scrolling works 4. There's almost certainly a better way to format JSON than "all on 1 line with spaces to make Ratatui wrapping work". But I think that works OK for now.
2025-06-03 14:29:26 -07:00
"unicode-segmentation",
fix: tighten up some logic around session timestamps and ids (#922) * update `SessionConfigured` event to include the UUID for the session * show the UUID in the Rust TUI * use local timestamps in log files instead of UTC * include timestamps in log file names for easier discovery
2025-05-13 19:22:16 -07:00
"uuid",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "color-eyre"
version = "0.6.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "55146f5e46f237f7423d74111267d4597b59b0dad0ffaf7303bce9945d843ad5"
dependencies = [
"backtrace",
"color-spantrace",
"eyre",
"indenter",
"once_cell",
feat: improve output of exec subcommand (#719)
2025-04-29 09:59:35 -07:00
"owo-colors 3.5.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tracing-error",
]
[[package]]
name = "color-spantrace"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cd6be1b2a7e382e2b98b43b2adcca6bb0e465af0bdd38123873ae61eb17a72c2"
dependencies = [
"once_cell",
feat: improve output of exec subcommand (#719)
2025-04-29 09:59:35 -07:00
"owo-colors 3.5.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"tracing-core",
"tracing-error",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "color_quant"
version = "1.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3d7b894f5411737b7867f4827955924d7c254fc9f4d91a6aad6b097804b1018b"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "colorchoice"
version = "1.0.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5b63caa9aa9397e2d9480a9b13673856c78d8ac123288526c37d7839f2a86990"
[[package]]
name = "compact_str"
version = "0.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3b79c4069c6cad78e2e0cdfcbd26275770669fb39fd308a752dc110e83b9af32"
dependencies = [
"castaway",
"cfg-if",
"itoa",
"rustversion",
"ryu",
"static_assertions",
]
[[package]]
name = "concurrent-queue"
version = "2.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4ca0197aee26d1ae37445ee532fefce43251d24cc7c166799f4d46817f1d3973"
dependencies = [
"crossbeam-utils",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "convert_case"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ec182b0ca2f35d8fc196cf3404988fd8b8c739a4d270ff118a398feb0cbec1ca"
dependencies = [
"unicode-segmentation",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "core-foundation"
version = "0.9.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "91e195e091a93c46f7102ec7818a2aa394e1e1771c3ab4825963fa03e45afb8f"
dependencies = [
"core-foundation-sys",
"libc",
]
[[package]]
name = "core-foundation-sys"
version = "0.8.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "773648b94d0e5d620f64f280777445740e61fe701025087ec8b57f45c791888b"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "crc32fast"
version = "1.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a97769d94ddab943e4510d138150169a2758b5ef3eb191a9ee688de3e23ef7b3"
dependencies = [
"cfg-if",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "crossbeam-channel"
version = "0.5.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "82b8f8f868b36967f9606790d1903570de9ceaf870a7bf9fbbd3016d636a2cb2"
dependencies = [
"crossbeam-utils",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "crossbeam-deque"
version = "0.8.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9dd111b7b7f7d55b72c0a6ae361660ee5853c9af73f70c3c2ef6858b950e2e51"
dependencies = [
"crossbeam-epoch",
"crossbeam-utils",
]
[[package]]
name = "crossbeam-epoch"
version = "0.9.18"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5b82ac4a3c2ca9c3460964f020e1402edd5753411d7737aa39c3714ad1b5420e"
dependencies = [
"crossbeam-utils",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "crossbeam-utils"
version = "0.8.21"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d0a5c400df2834b80a4c3327b3aad3a4c4cd4de0629063962b03235697506a28"
[[package]]
name = "crossterm"
version = "0.28.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "829d955a0bb380ef178a640b91779e3987da38c9aea133b20614cfed8cdea9c6"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"crossterm_winapi",
"mio",
"parking_lot",
"rustix 0.38.44",
"signal-hook",
"signal-hook-mio",
"winapi",
]
[[package]]
name = "crossterm_winapi"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "acdd7c62a3665c7f6830a51635d9ac9b23ed385797f70a83bb8bafe9c572ab2b"
dependencies = [
"winapi",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "crunchy"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "43da5946c66ffcc7745f48db692ffbb10a83bfe0afd96235c5c2a4fb23994929"
[[package]]
name = "ctor"
version = "0.1.26"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6d2301688392eb071b0bf1a37be05c469d3cc4dbbd95df672fe28ab021e6a096"
dependencies = [
"quote",
"syn 1.0.109",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "darling"
version = "0.20.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fc7f46116c46ff9ab3eb1597a45688b6715c6e628b5c133e288e709a29bcb4ee"
dependencies = [
"darling_core",
"darling_macro",
]
[[package]]
name = "darling_core"
version = "0.20.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0d00b9596d185e565c2207a0b01f8bd1a135483d02d9b7b0a54b11da8d53412e"
dependencies = [
"fnv",
"ident_case",
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"strsim 0.11.1",
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "darling_macro"
version = "0.20.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fc34b93ccb385b40dc71c6fceac4b2ad23662c7eeb248cf10d529b7e055b6ead"
dependencies = [
"darling_core",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "deadpool"
version = "0.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fb84100978c1c7b37f09ed3ce3e5f843af02c2a2c431bae5b19230dad2c1b490"
dependencies = [
"async-trait",
"deadpool-runtime",
"num_cpus",
"tokio",
]
[[package]]
name = "deadpool-runtime"
version = "0.1.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "092966b41edc516079bdf31ec78a2e0588d1d0c08f78b91d8307215928642b2b"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "debugserver-types"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2bf6834a70ed14e8e4e41882df27190bea150f1f6ecf461f1033f8739cd8af4a"
dependencies = [
"schemafy",
"serde",
"serde_json",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "deranged"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9c9e6a11ca8224451684bc0d7d5a7adbf8f2fd6887261a1cfc3c0432f9d4068e"
dependencies = [
"powerfmt",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"serde",
]
[[package]]
name = "derivative"
version = "2.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fcc3dd5e9e9c0b295d6e1e4d811fb6f157d5ffd784b8d202fc62eac8035a770b"
dependencies = [
"proc-macro2",
"quote",
"syn 1.0.109",
]
[[package]]
name = "derive_more"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4a9b99b9cbbe49445b21764dc0625032a89b145a2642e67603e1c936f5458d05"
dependencies = [
"derive_more-impl",
]
[[package]]
name = "derive_more-impl"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cb7330aeadfbe296029522e6c40f315320aba36fc43a5b3632f3795348f3bd22"
dependencies = [
"convert_case",
"proc-macro2",
"quote",
"syn 2.0.100",
"unicode-xid",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "diff"
version = "0.1.13"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56254986775e3233ffa9c4d7d3faaf6d36a2c09d30b20687e9f88bc8bafc16c8"
[[package]]
name = "difflib"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6184e33543162437515c2e2b48714794e37845ec9851711914eec9d308f6ebe8"
[[package]]
name = "dirs"
version = "6.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c3e8aa94d75141228480295a7d0e7feb620b1a5ad9f12bc40be62411e38cce4e"
dependencies = [
"dirs-sys",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "dirs-next"
version = "2.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b98cf8ebf19c3d1b223e151f99a4f9f0690dca41414773390fc824184ac833e1"
dependencies = [
"cfg-if",
"dirs-sys-next",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "dirs-sys"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e01a3366d27ee9890022452ee61b2b63a67e6f13f58900b651ff5665f0bb1fab"
dependencies = [
"libc",
"option-ext",
"redox_users 0.5.0",
"windows-sys 0.59.0",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "dirs-sys-next"
version = "0.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4ebda144c4fe02d1f7ea1a7d9641b6fc6b580adcfa024ae48797ecdeb6825b4d"
dependencies = [
"libc",
"redox_users 0.4.6",
"winapi",
]
[[package]]
name = "display_container"
version = "0.9.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0a110a75c96bedec8e65823dea00a1d710288b7a369d95fd8a0f5127639466fa"
dependencies = [
"either",
"indenter",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "displaydoc"
version = "0.2.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "97369cbbc041bc366949bc74d34658d6cda5621039731c6310521892a3a20ae0"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "doc-comment"
version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fea41bba32d969b513997752735605054bc0dfa92b4c56bf1189f2e174be7a10"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "dupe"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6ed2bc011db9c93fbc2b6cdb341a53737a55bafb46dbb74cf6764fc33a2fbf9c"
dependencies = [
"dupe_derive",
]
[[package]]
name = "dupe_derive"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "83e195b4945e88836d826124af44fdcb262ec01ef94d44f14f4fb5103f19892a"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: make Codex available as a tool when running it as an MCP server (#811) This PR replaces the placeholder `"echo"` tool call in the MCP server with a `"codex"` tool that calls Codex. Events such as `ExecApprovalRequest` and `ApplyPatchApprovalRequest` are not handled properly yet, but I have `approval_policy = "never"` set in my `~/.codex/config.toml` such that those codepaths are not exercised. The schema for this MPC tool is defined by a new `CodexToolCallParam` struct introduced in this PR. It is fairly similar to `ConfigOverrides`, as the param is used to help create the `Config` used to start the Codex session, though it also includes the `prompt` used to kick off the session. This PR also introduces the use of the third-party `schemars` crate to generate the JSON schema, which is verified in the `verify_codex_tool_json_schema()` unit test. Events that are dispatched during the Codex session are sent back to the MCP client as MCP notifications. This gives the client a way to monitor progress as the tool call itself may take minutes to complete depending on the complexity of the task requested by the user. In the video below, I launched the server via: ```shell mcp-server$ RUST_LOG=debug npx @modelcontextprotocol/inspector cargo run -- ``` In the video, you can see the flow of: * requesting the list of tools * choosing the **codex** tool * entering a value for **prompt** and then making the tool call Note that I left the other fields blank because when unspecified, the values in my `~/.codex/config.toml` were used: https://github.com/user-attachments/assets/1975058c-b004-43ef-8c8d-800a953b8192 Note that while using the inspector, I did run into https://github.com/modelcontextprotocol/inspector/issues/293, though the tip about ensuring I had only one instance of the **MCP Inspector** tab open in my browser seemed to fix things.
2025-05-05 07:16:19 -07:00
[[package]]
name = "dyn-clone"
version = "1.0.19"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1c7a8fb8a9fbf66c1f703fe16184d10ca0ee9d23be5b4436400408ba54a95005"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "either"
version = "1.15.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "ena"
version = "0.14.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3d248bdd43ce613d87415282f69b9bb99d947d290b10962dd6c56233312c2ad5"
dependencies = [
"log",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "encoding_rs"
version = "0.8.35"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "75030f3c4f45dafd7586dd6780965a8c7e8e285a5ecb86713e63a79c5b2766f3"
dependencies = [
"cfg-if",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "endian-type"
version = "0.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c34f04666d835ff5d62e058c3995147c06f42fe86ff053337632bca83e42702d"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "enumflags2"
version = "0.7.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ba2f4b465f5318854c6f8dd686ede6c0a9dc67d4b1ac241cf0eb51521a309147"
dependencies = [
"enumflags2_derive",
]
[[package]]
name = "enumflags2_derive"
version = "0.7.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fc4caf64a58d7a6d65ab00639b046ff54399a39f5f2554728895ace4b297cd79"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "env-flags"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dbfd0e7fc632dec5e6c9396a27bc9f9975b4e039720e1fd3e34021d3ce28c415"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "env_filter"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "186e05a59d4c50738528153b83b0b0194d3a29507dfec16eccd4b342903397d0"
dependencies = [
"log",
"regex",
]
[[package]]
name = "env_logger"
version = "0.11.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "13c863f0904021b108aa8b2f55046443e6b1ebde8fd4a15c399893aae4fa069f"
dependencies = [
"anstream",
"anstyle",
"env_filter",
"jiff",
"log",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "equivalent"
version = "1.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "erased-serde"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6c138974f9d5e7fe373eb04df7cae98833802ae4b11c24ac7039a21d5af4b26c"
dependencies = [
"serde",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "errno"
version = "0.3.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "976dd42dc7e85965fe702eb8164f21f450704bdde31faefd6471dba214cb594e"
dependencies = [
"libc",
"windows-sys 0.59.0",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "error-code"
version = "3.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a5d9305ccc6942a704f4335694ecd3de2ea531b114ac2d51f5f843750787a92f"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "event-listener"
version = "5.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3492acde4c3fc54c845eaab3eed8bd00c7a7d881f78bfc801e43a93dec1331ae"
dependencies = [
"concurrent-queue",
"parking",
"pin-project-lite",
]
[[package]]
name = "event-listener-strategy"
version = "0.5.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8be9f3dfaaffdae2972880079a491a1a8bb7cbed0b8dd7a347f668b4150a3b93"
dependencies = [
"event-listener",
"pin-project-lite",
]
[[package]]
name = "eventsource-stream"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "74fef4569247a5f429d9156b9d0a2599914385dd189c539334c625d8099d90ab"
dependencies = [
"futures-core",
"nom",
"pin-project-lite",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "exr"
version = "1.73.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f83197f59927b46c04a183a619b7c29df34e63e63c7869320862268c0ef687e0"
dependencies = [
"bit_field",
"half",
"lebe",
"miniz_oxide 0.8.8",
"rayon-core",
"smallvec",
"zune-inflate",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "eyre"
version = "0.6.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7cd915d99f24784cdc19fd37ef22b97e3ff0ae756c7e492e9fbfe897d61e2aec"
dependencies = [
"indenter",
"once_cell",
]
[[package]]
name = "fastrand"
version = "2.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "fd-lock"
version = "4.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0ce92ff622d6dadf7349484f42c93271a0d49b7cc4d466a936405bacbe10aa78"
dependencies = [
"cfg-if",
"rustix 1.0.5",
"windows-sys 0.59.0",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "fdeflate"
version = "0.3.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e6853b52649d4ac5c0bd02320cddc5ba956bdb407c4b75a2c6b75bf51500f8c"
dependencies = [
"simd-adler32",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "fixedbitset"
version = "0.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0ce7134b9999ecaf8bcd65542e436736ef32ddca1b3e06094cb6ec5755203b80"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "flate2"
version = "1.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7ced92e76e966ca2fd84c8f7aa01a4aea65b0eb6648d72f7c8f3e2764a67fece"
dependencies = [
"crc32fast",
"miniz_oxide 0.8.8",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "float-cmp"
version = "0.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b09cf3155332e944990140d967ff5eceb70df778b34f77d8075db46e4704e6d8"
dependencies = [
"num-traits",
]
[[package]]
name = "fnv"
version = "1.0.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
[[package]]
name = "foldhash"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2"
[[package]]
name = "foreign-types"
version = "0.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f6f339eb8adc052cd2ca78910fda869aefa38d22d5cb648e6485e4d3fc06f3b1"
dependencies = [
"foreign-types-shared",
]
[[package]]
name = "foreign-types-shared"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "00b0228411908ca8685dba7fc2cdd70ec9990a6e753e89b6ac91a84c40fbaf4b"
[[package]]
name = "form_urlencoded"
version = "1.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e13624c2627564efccf4934284bdd98cbaa14e79b0b5a141218e507b3a823456"
dependencies = [
"percent-encoding",
]
[[package]]
name = "fs-err"
version = "3.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1f89bda4c2a21204059a977ed3bfe746677dfd137b83c339e702b0ac91d482aa"
dependencies = [
"autocfg",
]
feat: record messages from user in ~/.codex/history.jsonl (#939) This is a large change to support a "history" feature like you would expect in a shell like Bash. History events are recorded in `$CODEX_HOME/history.jsonl`. Because it is a JSONL file, it is straightforward to append new entries (as opposed to the TypeScript file that uses `$CODEX_HOME/history.json`, so to be valid JSON, each new entry entails rewriting the entire file). Because it is possible for there to be multiple instances of Codex CLI writing to `history.jsonl` at once, we use advisory file locking when working with `history.jsonl` in `codex-rs/core/src/message_history.rs`. Because we believe history is a sufficiently useful feature, we enable it by default. Though to provide some safety, we set the file permissions of `history.jsonl` to be `o600` so that other users on the system cannot read the user's history. We do not yet support a default list of `SENSITIVE_PATTERNS` as the TypeScript CLI does: https://github.com/openai/codex/blob/3fdf9df1335ac9501e3fb0e61715359145711e8b/codex-cli/src/utils/storage/command-history.ts#L10-L17 We are going to take a more conservative approach to this list in the Rust CLI. For example, while `/\b[A-Za-z0-9-_]{20,}\b/` might exclude sensitive information like API tokens, it would also exclude valuable information such as references to Git commits. As noted in the updated documentation, users can opt-out of history by adding the following to `config.toml`: ```toml [history] persistence = "none" ``` Because `history.jsonl` could, in theory, be quite large, we take a[n arguably overly pedantic] approach in reading history entries into memory. Specifically, we start by telling the client the current number of entries in the history file (`history_entry_count`) as well as the inode (`history_log_id`) of `history.jsonl` (see the new fields on `SessionConfiguredEvent`). The client is responsible for keeping new entries in memory to create a "local history," but if the user hits up enough times to go "past" the end of local history, then the client should use the new `GetHistoryEntryRequest` in the protocol to fetch older entries. Specifically, it should pass the `history_log_id` it was given originally and work backwards from `history_entry_count`. (It should really fetch history in batches rather than one-at-a-time, but that is something we can improve upon in subsequent PRs.) The motivation behind this crazy scheme is that it is designed to defend against: * The `history.jsonl` being truncated during the session such that the index into the history is no longer consistent with what had been read up to that point. We do not yet have logic to enforce a `max_bytes` for `history.jsonl`, but once we do, we will aspire to implement it in a way that should result in a new inode for the file on most systems. * New items from concurrent Codex CLI sessions amending to the history. Because, in absence of truncation, `history.jsonl` is an append-only log, so long as the client reads backwards from `history_entry_count`, it should always get a consistent view of history. (That said, it will not be able to read _new_ commands from concurrent sessions, but perhaps we will introduce a `/` command to reload latest history or something down the road.) Admittedly, my testing of this feature thus far has been fairly light. I expect we will find bugs and introduce enhancements/fixes going forward.
2025-05-15 16:26:23 -07:00
[[package]]
name = "fs2"
version = "0.4.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9564fc758e15025b46aa6643b1b77d047d1a56a1aea6e01002ac0c7026876213"
dependencies = [
"libc",
"winapi",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "futures"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "65bc07b1a8bc7c85c5f2e110c476c7389b4554ba72af57d8445ea63a576b0876"
dependencies = [
"futures-channel",
"futures-core",
"futures-executor",
"futures-io",
"futures-sink",
"futures-task",
"futures-util",
]
[[package]]
name = "futures-channel"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2dff15bf788c671c1934e366d07e30c1814a8ef514e1af724a602e8a2fbe1b10"
dependencies = [
"futures-core",
"futures-sink",
]
[[package]]
name = "futures-core"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "05f29059c0c2090612e8d742178b0580d2dc940c837851ad723096f87af6663e"
[[package]]
name = "futures-executor"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e28d1d997f585e54aebc3f97d39e72338912123a67330d723fdbb564d646c9f"
dependencies = [
"futures-core",
"futures-task",
"futures-util",
]
[[package]]
name = "futures-io"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9e5c1b78ca4aae1ac06c48a526a655760685149f0d465d21f37abfe57ce075c6"
[[package]]
name = "futures-macro"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "162ee34ebcb7c64a8abebc059ce0fee27c2262618d7b60ed8faf72fef13c3650"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "futures-sink"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e575fab7d1e0dcb8d0c7bcf9a63ee213816ab51902e6d244a95819acacf1d4f7"
[[package]]
name = "futures-task"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f90f7dce0722e95104fcb095585910c0977252f286e354b5e3bd38902cd99988"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "futures-timer"
version = "3.0.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f288b0a4f20f9a56b5d1da57e2227c661b7b16168e2f72365f57b63326e29b24"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "futures-util"
version = "0.3.31"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9fa08315bb612088cc391249efdc3bc77536f16c91f6cf495e6fbe85b20a4a81"
dependencies = [
"futures-channel",
"futures-core",
"futures-io",
"futures-macro",
"futures-sink",
"futures-task",
"memchr",
"pin-project-lite",
"pin-utils",
"slab",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "fxhash"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c31b6d751ae2c7f11320402d34e41349dd1016f8d5d45e48c4312bc8625af50c"
dependencies = [
"byteorder",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "getopts"
version = "0.2.21"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "14dbbfd5c71d70241ecf9e6f13737f7b5ce823821063188d7e46c41d371eebd5"
dependencies = [
"unicode-width 0.1.14",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "getrandom"
version = "0.2.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "335ff9f135e4384c8150d6f27c6daed433577f86b4750418338c01a1a2528592"
dependencies = [
"cfg-if",
"libc",
"wasi 0.11.0+wasi-snapshot-preview1",
]
[[package]]
name = "getrandom"
version = "0.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "73fea8450eea4bac3940448fb7ae50d91f034f941199fcd9d909a5a07aa455f0"
dependencies = [
"cfg-if",
"libc",
"r-efi",
"wasi 0.14.2+wasi-0.2.4",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "gif"
version = "0.13.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3fb2d69b19215e18bb912fa30f7ce15846e301408695e44e0ef719f1da9e19f2"
dependencies = [
"color_quant",
"weezl",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "gimli"
version = "0.28.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4271d37baee1b8c7e4b708028c57d816cf9d2434acb33a549475f78c181f6253"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "glob"
version = "0.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a8d1add55171497b4705a648c6b583acafb01d58050a51727785f0b2c8e0a2b2"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "h2"
version = "0.4.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "75249d144030531f8dee69fe9cea04d3edf809a017ae445e2abdff6629e86633"
dependencies = [
"atomic-waker",
"bytes",
"fnv",
"futures-core",
"futures-sink",
"http",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"indexmap 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"slab",
"tokio",
"tokio-util",
"tracing",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "half"
version = "2.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "459196ed295495a68f7d7fe1d84f6c4b7ff0e21fe3017b2f283c6fac3ad803c9"
dependencies = [
"cfg-if",
"crunchy",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "hashbrown"
version = "0.12.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8a9ee70c43aaf417c914396645a0fa852624801b24ebb7ae78fe8272889ac888"
[[package]]
name = "hashbrown"
version = "0.14.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e5274423e17b7c9fc20b6e7e208532f9b19825d82dfd615708b70edd83df41f1"
dependencies = [
"ahash",
"allocator-api2",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "hashbrown"
version = "0.15.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bf151400ff0baff5465007dd2f3e717f3fe502074ca563069ce3a6629d07b289"
dependencies = [
"allocator-api2",
"equivalent",
"foldhash",
]
[[package]]
name = "heck"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2304e00983f87ffb38b55b444b5e3b60a884b5d30c0fca7d82fe33449bbe55ea"
[[package]]
name = "hermit-abi"
version = "0.3.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d231dfb89cfffdbc30e7fc41579ed6066ad03abda9e567ccafae602b97ec5024"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "hermit-abi"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fbd780fe5cc30f81464441920d82ac8740e2e46b29a6fad543ddd075229ce37e"
[[package]]
name = "hex"
version = "0.4.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7f24254aa9a54b5c858eaee2f5bccdb46aaf0e486a595ed5fd8f86ba55232a70"
[[package]]
name = "home"
version = "0.5.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "589533453244b0995c858700322199b2becb13b627df2851f64a2775d024abcf"
dependencies = [
"windows-sys 0.59.0",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "http"
version = "1.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f4a85d31aea989eead29a3aaf9e1115a180df8282431156e533de47660892565"
dependencies = [
"bytes",
"fnv",
"itoa",
]
[[package]]
name = "http-body"
version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1efedce1fb8e6913f23e0c92de8e62cd5b772a67e7b3946df930a62566c93184"
dependencies = [
"bytes",
"http",
]
[[package]]
name = "http-body-util"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b021d93e26becf5dc7e1b75b1bed1fd93124b374ceb73f43d4d4eafec896a64a"
dependencies = [
"bytes",
"futures-core",
"http",
"http-body",
"pin-project-lite",
]
[[package]]
name = "httparse"
version = "1.10.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6dbf3de79e51f3d586ab4cb9d5c3e2c14aa28ed23d180cf89b4df0454a69cc87"
[[package]]
name = "httpdate"
version = "1.0.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "df3b46402a9d5adb4c86a0cf463f42e19994e3ee891101b1841f30a545cb49a9"
[[package]]
name = "hyper"
version = "1.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cc2b571658e38e0c01b1fdca3bbbe93c00d3d71693ff2770043f8c29bc7d6f80"
dependencies = [
"bytes",
"futures-channel",
"futures-util",
"h2",
"http",
"http-body",
"httparse",
"httpdate",
"itoa",
"pin-project-lite",
"smallvec",
"tokio",
"want",
]
[[package]]
name = "hyper-rustls"
version = "0.27.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2d191583f3da1305256f22463b9bb0471acad48a4e534a5218b9963e9c1f59b2"
dependencies = [
"futures-util",
"http",
"hyper",
"hyper-util",
"rustls",
"rustls-pki-types",
"tokio",
"tokio-rustls",
"tower-service",
]
[[package]]
name = "hyper-tls"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "70206fc6890eaca9fde8a0bf71caa2ddfc9fe045ac9e5c70df101a7dbde866e0"
dependencies = [
"bytes",
"http-body-util",
"hyper",
"hyper-util",
"native-tls",
"tokio",
"tokio-native-tls",
"tower-service",
]
[[package]]
name = "hyper-util"
version = "0.1.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "497bbc33a26fdd4af9ed9c70d63f61cf56a938375fbb32df34db9b1cd6d643f2"
dependencies = [
"bytes",
"futures-channel",
"futures-util",
"http",
"http-body",
"hyper",
"libc",
"pin-project-lite",
"socket2",
"tokio",
"tower-service",
"tracing",
]
[[package]]
name = "iana-time-zone"
version = "0.1.63"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b0c919e5debc312ad217002b8048a17b7d83f80703865bbfcfebb0458b0b27d8"
dependencies = [
"android_system_properties",
"core-foundation-sys",
"iana-time-zone-haiku",
"js-sys",
"log",
"wasm-bindgen",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"windows-core 0.61.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "iana-time-zone-haiku"
version = "0.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f31827a206f56af32e590ba56d5d2d085f558508192593743f16b2306495269f"
dependencies = [
"cc",
]
[[package]]
name = "icu_collections"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "db2fa452206ebee18c4b5c2274dbf1de17008e874b4dc4f0aea9d01ca79e4526"
dependencies = [
"displaydoc",
"yoke",
"zerofrom",
"zerovec",
]
[[package]]
name = "icu_locid"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "13acbb8371917fc971be86fc8057c41a64b521c184808a698c02acc242dbf637"
dependencies = [
"displaydoc",
"litemap",
"tinystr",
"writeable",
"zerovec",
]
[[package]]
name = "icu_locid_transform"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "01d11ac35de8e40fdeda00d9e1e9d92525f3f9d887cdd7aa81d727596788b54e"
dependencies = [
"displaydoc",
"icu_locid",
"icu_locid_transform_data",
"icu_provider",
"tinystr",
"zerovec",
]
[[package]]
name = "icu_locid_transform_data"
version = "1.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7515e6d781098bf9f7205ab3fc7e9709d34554ae0b21ddbcb5febfa4bc7df11d"
[[package]]
name = "icu_normalizer"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "19ce3e0da2ec68599d193c93d088142efd7f9c5d6fc9b803774855747dc6a84f"
dependencies = [
"displaydoc",
"icu_collections",
"icu_normalizer_data",
"icu_properties",
"icu_provider",
"smallvec",
"utf16_iter",
"utf8_iter",
"write16",
"zerovec",
]
[[package]]
name = "icu_normalizer_data"
version = "1.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c5e8338228bdc8ab83303f16b797e177953730f601a96c25d10cb3ab0daa0cb7"
[[package]]
name = "icu_properties"
version = "1.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "93d6020766cfc6302c15dbbc9c8778c37e62c14427cb7f6e601d849e092aeef5"
dependencies = [
"displaydoc",
"icu_collections",
"icu_locid_transform",
"icu_properties_data",
"icu_provider",
"tinystr",
"zerovec",
]
[[package]]
name = "icu_properties_data"
version = "1.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "85fb8799753b75aee8d2a21d7c14d9f38921b54b3dbda10f5a3c7a7b82dba5e2"
[[package]]
name = "icu_provider"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6ed421c8a8ef78d3e2dbc98a973be2f3770cb42b606e3ab18d6237c4dfde68d9"
dependencies = [
"displaydoc",
"icu_locid",
"icu_provider_macros",
"stable_deref_trait",
"tinystr",
"writeable",
"yoke",
"zerofrom",
"zerovec",
]
[[package]]
name = "icu_provider_macros"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1ec89e9337638ecdc08744df490b221a7399bf8d164eb52a665454e60e075ad6"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "icy_sixel"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ccc0a9c4770bc47b0a933256a496cfb8b6531f753ea9bccb19c6dff0ff7273fc"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "ident_case"
version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b9e0384b61958566e926dc50660321d12159025e767c18e043daf26b70104c39"
[[package]]
name = "idna"
version = "1.0.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "686f825264d630750a544639377bae737628043f20d38bbc029e8f29ea968a7e"
dependencies = [
"idna_adapter",
"smallvec",
"utf8_iter",
]
[[package]]
name = "idna_adapter"
version = "1.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "daca1df1c957320b2cf139ac61e7bd64fed304c5040df000a745aa1de3b4ef71"
dependencies = [
"icu_normalizer",
"icu_properties",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "image"
version = "0.25.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "db35664ce6b9810857a38a906215e75a9c879f0696556a39f59c62829710251a"
dependencies = [
"bytemuck",
"byteorder-lite",
"color_quant",
"exr",
"gif",
"image-webp",
"num-traits",
"png",
"qoi",
"ravif",
"rayon",
"rgb",
"tiff",
"zune-core",
"zune-jpeg",
]
[[package]]
name = "image-webp"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b77d01e822461baa8409e156015a1d91735549f0f2c17691bd2d996bef238f7f"
dependencies = [
"byteorder-lite",
"quick-error",
]
[[package]]
name = "imgref"
version = "1.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d0263a3d970d5c054ed9312c0057b4f3bde9c0b33836d3637361d4a9e6e7a408"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "indenter"
version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ce23b50ad8242c51a442f3ff322d56b02f08852c77e4c0b4d3fd684abc89c683"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "indexmap"
version = "1.9.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bd070e393353796e801d209ad339e89596eb4c8d430d18ede6a1cced8fafbd99"
dependencies = [
"autocfg",
"hashbrown 0.12.3",
"serde",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "indexmap"
version = "2.9.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cea70ddb795996207ad57735b50c5982d8844f38ba9ee5f1aedcfb708a2aa11e"
dependencies = [
"equivalent",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"hashbrown 0.15.2",
"serde",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "indoc"
version = "2.0.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f4c7245a08504955605670dbf141fceab975f15ca21570696aebe9d2e71576bd"
[[package]]
name = "instability"
version = "0.3.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0bf9fed6d91cfb734e7476a06bde8300a1b94e217e1b523b6f0cd1a01998c71d"
dependencies = [
"darling",
"indoc",
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "interpolate_name"
version = "0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c34819042dc3d3971c46c2190835914dfbe0c3c13f61449b2997f4e9722dfa60"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "inventory"
version = "0.3.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ab08d7cd2c5897f2c949e5383ea7c7db03fb19130ffcfbf7eda795137ae3cb83"
dependencies = [
"rustversion",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "ipnet"
version = "2.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "469fb0b9cefa57e3ef31275ee7cacb78f2fdca44e4765491884a2b119d4eb130"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "is-terminal"
version = "0.4.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e04d7f318608d35d4b61ddd75cbdaee86b023ebe2bd5a66ee0915f0bf93095a9"
dependencies = [
"hermit-abi 0.5.0",
"libc",
"windows-sys 0.59.0",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "is_terminal_polyfill"
version = "1.70.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7943c866cc5cd64cbc25b2e01621d07fa8eb2a1a23160ee81ce38704e97b8ecf"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "itertools"
version = "0.10.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b0fd2260e829bddf4cb6ea802289de2f86d6a7a690192fbe91b3f46e0f2c8473"
dependencies = [
"either",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "itertools"
version = "0.12.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ba291022dbbd398a455acf126c1e341954079855bc60dfdda641363bd6922569"
dependencies = [
"either",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "itertools"
version = "0.13.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "413ee7dfc52ee1a4949ceeb7dbc8a33f2d6c088194d9f922fb8318faf1f01186"
dependencies = [
"either",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "itertools"
version = "0.14.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2b192c782037fadd9cfa75548310488aabdbf3d2da73885b31bd0abd03351285"
dependencies = [
"either",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "itoa"
version = "1.0.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4a5f13b858c8d314ee3e8f639011f7ccefe71f97f96e50151fb991f267928e2c"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "jiff"
version = "0.2.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5a064218214dc6a10fbae5ec5fa888d80c45d611aba169222fc272072bf7aef6"
dependencies = [
"jiff-static",
"log",
"portable-atomic",
"portable-atomic-util",
"serde",
]
[[package]]
name = "jiff-static"
version = "0.2.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "199b7932d97e325aff3a7030e141eafe7f2c6268e1d1b24859b753a627f45254"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "jobserver"
version = "0.1.33"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "38f262f097c174adebe41eb73d66ae9c06b2844fb0da69969647bbddd9b0538a"
dependencies = [
"getrandom 0.3.2",
"libc",
]
[[package]]
name = "jpeg-decoder"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f5d4a7da358eff58addd2877a45865158f0d78c911d43a5784ceb7bbf52833b0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "js-sys"
version = "0.3.77"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1cfaf33c695fc6e08064efbc1f72ec937429614f25eef83af942d0e227c3a28f"
dependencies = [
"once_cell",
"wasm-bindgen",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "lalrpop"
version = "0.19.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0a1cbf952127589f2851ab2046af368fd20645491bb4b376f04b7f94d7a9837b"
dependencies = [
"ascii-canvas",
"bit-set",
"diff",
"ena",
"is-terminal",
"itertools 0.10.5",
"lalrpop-util",
"petgraph",
"regex",
"regex-syntax 0.6.29",
"string_cache",
"term",
"tiny-keccak",
"unicode-xid",
]
[[package]]
name = "lalrpop-util"
version = "0.19.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d3c48237b9604c5a4702de6b824e02006c3214327564636aef27c1028a8fa0ed"
dependencies = [
"regex",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "landlock"
version = "0.4.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "18738c5d4c7fae6727a96adb94722ef7ce82f3eafea0a11777e258a93816537e"
dependencies = [
"enumflags2",
"libc",
"thiserror 1.0.69",
]
[[package]]
name = "lazy_static"
version = "1.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bbd2bcb4c963f2ddae06a2efc7e9f3591312473c50c6685e1f298068316e66fe"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "lebe"
version = "0.5.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "03087c2bad5e1034e8cace5926dec053fb3790248370865f5117a7d0213354c8"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "libc"
version = "0.2.172"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d750af042f7ef4f724306de029d18836c26c1765a54a6a3f094cbd23a7267ffa"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "libfuzzer-sys"
version = "0.4.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cf78f52d400cf2d84a3a973a78a592b4adc535739e0a5597a0da6f0c357adc75"
dependencies = [
"arbitrary",
"cc",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "libredox"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c0ff37bd590ca25063e35af745c343cb7a0271906fb7b37e4813e8f79f00268d"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"libc",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "linked-hash-map"
version = "0.5.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0717cef1bc8b636c6e1c1bbdefc09e6322da8a9321966e8928ef80d20f7f770f"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "linux-raw-sys"
version = "0.4.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d26c52dbd32dccf2d10cac7725f8eae5296885fb5703b261f7d0a0739ec807ab"
[[package]]
name = "linux-raw-sys"
version = "0.9.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cd945864f07fe9f5371a27ad7b52a172b4b499999f1d97574c9fa68373937e12"
[[package]]
name = "litemap"
version = "0.7.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "23fb14cb19457329c82206317a5663005a4d404783dc74f4252769b0d5f42856"
[[package]]
name = "lock_api"
version = "0.4.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "07af8b9cdd281b7915f413fa73f29ebd5d55d0d3f0155584dade1ff18cea1b17"
dependencies = [
"autocfg",
"scopeguard",
]
[[package]]
name = "log"
version = "0.4.27"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "13dc2df351e3202783a1fe0d44375f7295ffb4049267b0f3018346dc122a1d94"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "logos"
version = "0.12.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bf8b031682c67a8e3d5446840f9573eb7fe26efe7ec8d195c9ac4c0647c502f1"
dependencies = [
"logos-derive",
]
[[package]]
name = "logos-derive"
version = "0.12.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a1d849148dbaf9661a6151d1ca82b13bb4c4c128146a88d05253b38d4e2f496c"
dependencies = [
"beef",
"fnv",
"proc-macro2",
"quote",
"regex-syntax 0.6.29",
"syn 1.0.109",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "loop9"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0fae87c125b03c1d2c0150c90365d7d6bcc53fb73a9acaef207d2d065860f062"
dependencies = [
"imgref",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "lru"
version = "0.12.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "234cf4f4a04dc1f57e24b96cc0cd600cf2af460d4161ac5ecdd0af8e1f3b2a38"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"hashbrown 0.15.2",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "lsp-types"
version = "0.94.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c66bfd44a06ae10647fe3f8214762e9369fd4248df1350924b4ef9e770a85ea1"
dependencies = [
"bitflags 1.3.2",
"serde",
"serde_json",
"serde_repr",
"url",
]
[[package]]
name = "maplit"
version = "1.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3e2e65a1a2e43cfcb47a895c4c8b10d1f4a61097f9f254f183aee60cad9c651d"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "matchers"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8263075bb86c5a1b1427b5ae862e8889656f126e9f77c484496e8b47cf5c5558"
dependencies = [
"regex-automata 0.1.10",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "maybe-rayon"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8ea1f30cedd69f0a2954655f7188c6a834246d2bcf1e315e2ac40c4b24dc9519"
dependencies = [
"cfg-if",
"rayon",
]
feat: introduce mcp-types crate (#787) This adds our own `mcp-types` crate to our Cargo workspace. We vendor in the [`2025-03-26/schema.json`](https://github.com/modelcontextprotocol/modelcontextprotocol/blob/05f204513641c05bd78d056791af99c6c84520fa/schema/2025-03-26/schema.json) from the MCP repo and introduce a `generate_mcp_types.py` script to codegen the `lib.rs` from the JSON schema. Test coverage is currently light, but I plan to refine things as we start making use of this crate. And yes, I am aware that https://github.com/modelcontextprotocol/rust-sdk exists, though the published https://crates.io/crates/rmcp appears to be a competing effort. While things are up in the air, it seems better for us to control our own version of this code. Incidentally, Codex did a lot of the work for this PR. I told it to never edit `lib.rs` directly and instead to update `generate_mcp_types.py` and then re-run it to update `lib.rs`. It followed these instructions and once things were working end-to-end, I iteratively asked for changes to the tests until the API looked reasonable (and the code worked). Codex was responsible for figuring out what to do to `generate_mcp_types.py` to achieve the requested test/API changes.
2025-05-02 13:33:14 -07:00
[[package]]
name = "mcp-types"
Update submodules version to come from the workspace (#850) Tie the version of submodules to the workspace version.
2025-05-07 10:08:06 -07:00
version = "0.0.0"
feat: introduce mcp-types crate (#787) This adds our own `mcp-types` crate to our Cargo workspace. We vendor in the [`2025-03-26/schema.json`](https://github.com/modelcontextprotocol/modelcontextprotocol/blob/05f204513641c05bd78d056791af99c6c84520fa/schema/2025-03-26/schema.json) from the MCP repo and introduce a `generate_mcp_types.py` script to codegen the `lib.rs` from the JSON schema. Test coverage is currently light, but I plan to refine things as we start making use of this crate. And yes, I am aware that https://github.com/modelcontextprotocol/rust-sdk exists, though the published https://crates.io/crates/rmcp appears to be a competing effort. While things are up in the air, it seems better for us to control our own version of this code. Incidentally, Codex did a lot of the work for this PR. I told it to never edit `lib.rs` directly and instead to update `generate_mcp_types.py` and then re-run it to update `lib.rs`. It followed these instructions and once things were working end-to-end, I iteratively asked for changes to the tests until the API looked reasonable (and the code worked). Codex was responsible for figuring out what to do to `generate_mcp_types.py` to achieve the requested test/API changes.
2025-05-02 13:33:14 -07:00
dependencies = [
"serde",
"serde_json",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "memchr"
version = "2.7.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "78ca9ab1a0babb1e7d5695e3530886289c18cf2f87ec19a575a0abdce112e3a3"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "memoffset"
version = "0.6.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5aa361d4faea93603064a027415f07bd8e1d5c88c9fbf68bf56a285428fd79ce"
dependencies = [
"autocfg",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "mime"
version = "0.3.17"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6877bb514081ee2a7ff5ef9de3281f14a4dd4bceac4c09388074a6b5df8a139a"
[[package]]
name = "mime_guess"
version = "2.0.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f7c44f8e672c00fe5308fa235f821cb4198414e1c77935c1ab6948d3fd78550e"
dependencies = [
"mime",
"unicase",
]
[[package]]
name = "minimal-lexical"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
[[package]]
name = "miniz_oxide"
version = "0.7.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b8a240ddb74feaf34a79a7add65a741f3167852fba007066dcac1ca548d89c08"
dependencies = [
"adler",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "miniz_oxide"
version = "0.8.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3be647b768db090acb35d5ec5db2b0e1f1de11133ca123b9eacf5137868f892a"
dependencies = [
"adler2",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"simd-adler32",
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "mio"
version = "1.0.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2886843bf800fba2e3377cff24abf6379b4c4d5c6681eaf9ea5b0d15090450bd"
dependencies = [
"libc",
"log",
"wasi 0.11.0+wasi-snapshot-preview1",
"windows-sys 0.52.0",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "multimap"
version = "0.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "defc4c55412d89136f966bbb339008b474350e5e6e78d2714439c386b3137a03"
dependencies = [
"serde",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "native-tls"
version = "0.2.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "87de3442987e9dbec73158d5c715e7ad9072fda936bb03d19d7fa10e00520f0e"
dependencies = [
"libc",
"log",
"openssl",
"openssl-probe",
"openssl-sys",
"schannel",
"security-framework",
"security-framework-sys",
"tempfile",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "new_debug_unreachable"
version = "1.0.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "650eef8c711430f1a879fdd01d4745a7deea475becfb90269c06775983bbf086"
[[package]]
name = "nibble_vec"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "77a5d83df9f36fe23f0c3648c6bbb8b0298bb5f1939c8f2704431371f4b84d43"
dependencies = [
"smallvec",
]
[[package]]
name = "nix"
version = "0.28.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ab2156c4fce2f8df6c499cc1c763e4394b7482525bf2a9701c9d79d215f519e4"
dependencies = [
"bitflags 2.9.0",
"cfg-if",
"cfg_aliases",
"libc",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "nom"
version = "7.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
dependencies = [
"memchr",
"minimal-lexical",
]
[[package]]
name = "nom_locate"
version = "4.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e3c83c053b0713da60c5b8de47fe8e494fe3ece5267b2f23090a07a053ba8f3"
dependencies = [
"bytecount",
"memchr",
"nom",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "noop_proc_macro"
version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0676bb32a98c1a483ce53e500a81ad9c3d5b3f7c920c28c24e9cb0980d0b5bc8"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "normalize-line-endings"
version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "61807f77802ff30975e01f4f071c8ba10c022052f98b3294119f3e615d13e5be"
[[package]]
name = "nu-ansi-term"
version = "0.46.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "77a8165726e8236064dbb45459242600304b42a5ea24ee2948e18e023bf7ba84"
dependencies = [
"overload",
"winapi",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "num-bigint"
version = "0.4.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a5e44f723f1133c9deac646763579fdb3ac745e418f2a7af9cd0c431da1f20b9"
dependencies = [
"num-integer",
"num-traits",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "num-conv"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "51d515d32fb182ee37cda2ccdcb92950d6a3c2893aa280e540671c2cd0f3b1d9"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "num-derive"
version = "0.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ed3955f1a9c7c0c15e092f9c887db08b1fc683305fdf6eb6684f22555355e202"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "num-integer"
version = "0.1.46"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7969661fd2958a5cb096e56c8e1ad0444ac2bbcd0061bd28660485a44879858f"
dependencies = [
"num-traits",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "num-rational"
version = "0.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f83d14da390562dca69fc84082e73e548e1ad308d24accdedd2720017cb37824"
dependencies = [
"num-bigint",
"num-integer",
"num-traits",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "num-traits"
version = "0.2.19"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "071dfc062690e90b734c0b2273ce72ad0ffa95f0c74596bc250dcfd960262841"
dependencies = [
"autocfg",
]
[[package]]
name = "num_cpus"
version = "1.16.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4161fcb6d602d4d2081af7c3a45852d875a03dd337a6bfdd6e06407b61342a43"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"hermit-abi 0.3.9",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"libc",
]
fix: tighten up some logic around session timestamps and ids (#922) * update `SessionConfigured` event to include the UUID for the session * show the UUID in the Rust TUI * use local timestamps in log files instead of UTC * include timestamps in log file names for easier discovery
2025-05-13 19:22:16 -07:00
[[package]]
name = "num_threads"
version = "0.1.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5c7398b9c8b70908f6371f47ed36737907c87c52af34c268fed0bf0ceb92ead9"
dependencies = [
"libc",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "object"
version = "0.32.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a6a622008b6e321afc04970976f62ee297fdbaa6f95318ca343e3eebb9648441"
dependencies = [
"memchr",
]
[[package]]
name = "once_cell"
version = "1.21.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "onig"
version = "6.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8c4b31c8722ad9171c6d77d3557db078cab2bd50afcc9d09c8b315c59df8ca4f"
dependencies = [
"bitflags 1.3.2",
"libc",
"once_cell",
"onig_sys",
]
[[package]]
name = "onig_sys"
version = "69.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7b829e3d7e9cc74c7e315ee8edb185bf4190da5acde74afd7fc59c35b1f086e7"
dependencies = [
"cc",
"pkg-config",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "openssl"
version = "0.10.72"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fedfea7d58a1f73118430a55da6a286e7b044961736ce96a16a17068ea25e5da"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"cfg-if",
"foreign-types",
"libc",
"once_cell",
"openssl-macros",
"openssl-sys",
]
[[package]]
name = "openssl-macros"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a948666b637a0f465e8564c73e89d4dde00d72d4d473cc972f390fc3dcee7d9c"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "openssl-probe"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d05e27ee213611ffe7d6348b942e8f942b37114c00cc03cec254295a4a17852e"
[[package]]
name = "openssl-src"
version = "300.5.0+3.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e8ce546f549326b0e6052b649198487d91320875da901e7bd11a06d1ee3f9c2f"
dependencies = [
"cc",
]
[[package]]
name = "openssl-sys"
version = "0.9.107"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8288979acd84749c744a9014b4382d42b8f7b2592847b5afb2ed29e5d16ede07"
dependencies = [
"cc",
"libc",
"openssl-src",
"pkg-config",
"vcpkg",
]
[[package]]
name = "option-ext"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "04744f49eae99ab78e0d5c0b603ab218f515ea8cfe5a456d7629ad883a3b6e7d"
[[package]]
name = "overload"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b15813163c1d831bf4a13c3610c05c0d03b39feb07f7e09fa234dac9b15aaf39"
feat: improve output of exec subcommand (#719)
2025-04-29 09:59:35 -07:00
[[package]]
name = "owo-colors"
version = "3.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c1b04fb49957986fdce4d6ee7a65027d55d4b6d2265e5848bbb507b58ccfdb6f"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "owo-colors"
version = "4.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1036865bb9422d3300cf723f657c2851d0e9ab12567854b1f4eba3d77decf564"
[[package]]
name = "parking"
version = "2.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f38d5652c16fde515bb1ecef450ab0f6a219d619a7274976324d5e377f7dceba"
[[package]]
name = "parking_lot"
version = "0.12.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f1bf18183cf54e8d6059647fc3063646a1801cf30896933ec2311622cc4b9a27"
dependencies = [
"lock_api",
"parking_lot_core",
]
[[package]]
name = "parking_lot_core"
version = "0.9.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e401f977ab385c9e4e3ab30627d6f26d00e2c73eef317493c4ec6d468726cf8"
dependencies = [
"cfg-if",
"libc",
fix: remove dependency on expanduser crate (#667) In putting up https://github.com/openai/codex/pull/665, I discovered that the `expanduser` crate does not compile on Windows. Looking into it, we do not seem to need it because we were only using it with a value that was passed in via a command-line flag, so the shell expands `~` for us before we see it, anyway. (I changed the type in `Cli` from `String` to `PathBuf`, to boot.) If we do need this sort of functionality in the future, https://docs.rs/shellexpand/latest/shellexpand/fn.tilde.html seems promising.
2025-04-25 14:20:21 -07:00
"redox_syscall",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"smallvec",
"windows-targets 0.52.6",
]
[[package]]
name = "paste"
version = "1.0.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "57c0d7b74b563b49d38dae00a0c37d4d6de9b432382b2892f0574ddcae73fd0a"
[[package]]
name = "patch"
version = "0.7.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "15c07fdcdd8b05bdcf2a25bc195b6c34cbd52762ada9dba88bf81e7686d14e7a"
dependencies = [
"chrono",
"nom",
"nom_locate",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "path-absolutize"
version = "3.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e4af381fe79fa195b4909485d99f73a80792331df0625188e707854f0b3383f5"
dependencies = [
"path-dedot",
]
feat: add support for file_opener option in Rust, similiar to #911 (#957) This ports the enhancement introduced in https://github.com/openai/codex/pull/911 (and the fixes in https://github.com/openai/codex/pull/919) for the TypeScript CLI to the Rust one.
2025-05-16 11:33:08 -07:00
[[package]]
name = "path-clean"
version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "17359afc20d7ab31fdb42bb844c8b3bb1dabd7dcf7e68428492da7f16966fcef"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "path-dedot"
version = "3.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "07ba0ad7e047712414213ff67533e6dd477af0a4e1d14fb52343e53d30ea9397"
dependencies = [
"once_cell",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "percent-encoding"
version = "2.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "petgraph"
version = "0.6.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b4c5cc86750666a3ed20bdaf5ca2a0344f9c67674cae0515bec2da16fbaa47db"
dependencies = [
"fixedbitset",
"indexmap 2.9.0",
]
[[package]]
name = "phf_shared"
version = "0.11.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "67eabc2ef2a60eb7faa00097bd1ffdb5bd28e62bf39990626a582201b7a754e5"
dependencies = [
"siphasher",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "pin-project-lite"
version = "0.2.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3b3cff922bd51709b605d9ead9aa71031d81447142d828eb4a6eba76fe619f9b"
[[package]]
name = "pin-utils"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184"
[[package]]
name = "pkg-config"
version = "0.3.32"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7edddbd0b52d732b21ad9a5fab5c704c14cd949e5e9a1ec5929a24fded1b904c"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "plist"
version = "1.7.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "eac26e981c03a6e53e0aee43c113e3202f5581d5360dae7bd2c70e800dd0451d"
dependencies = [
"base64 0.22.1",
"indexmap 2.9.0",
"quick-xml",
"serde",
"time",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "png"
version = "0.17.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "82151a2fc869e011c153adc57cf2789ccb8d9906ce52c0b39a6b5697749d7526"
dependencies = [
"bitflags 1.3.2",
"crc32fast",
"fdeflate",
"flate2",
"miniz_oxide 0.8.8",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "portable-atomic"
version = "1.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "350e9b48cbc6b0e028b0473b114454c6316e57336ee184ceab6e53f72c178b3e"
[[package]]
name = "portable-atomic-util"
version = "0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d8a2f0d8d040d7848a709caf78912debcc3f33ee4b3cac47d73d1e1069e83507"
dependencies = [
"portable-atomic",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "powerfmt"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "439ee305def115ba05938db6eb1644ff94165c5ab5e9420d1c1bcedbba909391"
[[package]]
name = "ppv-lite86"
version = "0.2.21"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "85eae3c4ed2f50dcfe72643da4befc30deadb458a9b590d720cde2f2b1e97da9"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"zerocopy 0.8.24",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "precomputed-hash"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "925383efa346730478fb4838dbe9137d2a47675ad789c546d150a6e1dd4ab31c"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "predicates"
version = "3.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a5d19ee57562043d37e82899fade9a22ebab7be9cef5026b07fda9cdd4293573"
dependencies = [
"anstyle",
"difflib",
"float-cmp",
"normalize-line-endings",
"predicates-core",
"regex",
]
[[package]]
name = "predicates-core"
version = "1.0.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "727e462b119fe9c93fd0eb1429a5f7647394014cf3c04ab2c0350eeb09095ffa"
[[package]]
name = "predicates-tree"
version = "1.0.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "72dd2d6d381dfb73a193c7fca536518d7caee39fc8503f74e7dc0be0531b425c"
dependencies = [
"predicates-core",
"termtree",
]
[[package]]
name = "pretty_assertions"
version = "1.4.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3ae130e2f271fbc2ac3a40fb1d07180839cdbbe443c7a27e1e3c13c5cac0116d"
dependencies = [
"diff",
"yansi",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "proc-macro-crate"
version = "3.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "edce586971a4dfaa28950c6f18ed55e0406c1ab88bbce2c6f6293a7aaba73d35"
dependencies = [
"toml_edit",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "proc-macro2"
version = "1.0.95"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "02b3e5e68a3a1a02aad3ec490a98007cbc13c37cbe84a3cd7b8e406d76e7f778"
dependencies = [
"unicode-ident",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "profiling"
version = "1.0.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "afbdc74edc00b6f6a218ca6a5364d6226a259d4b8ea1af4a0ea063f27e179f4d"
dependencies = [
"profiling-procmacros",
]
[[package]]
name = "profiling-procmacros"
version = "1.0.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a65f2e60fbf1063868558d69c6beacf412dc755f9fc020f514b7955fc914fe30"
dependencies = [
"quote",
"syn 2.0.100",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "pulldown-cmark"
version = "0.13.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e8bbe1a966bd2f362681a44f6edce3c2310ac21e4d5067a6e7ec396297a6ea0"
dependencies = [
"bitflags 2.9.0",
"getopts",
"memchr",
"pulldown-cmark-escape",
"unicase",
]
[[package]]
name = "pulldown-cmark-escape"
version = "0.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "007d8adb5ddab6f8e3f491ac63566a7d5002cc7ed73901f72057943fa71ae1ae"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "qoi"
version = "0.4.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7f6d64c71eb498fe9eae14ce4ec935c555749aef511cca85b5568910d6e48001"
dependencies = [
"bytemuck",
]
[[package]]
name = "quick-error"
version = "2.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a993555f31e5a609f617c12db6250dedcac1b0a85076912c436e6fc9b2c8e6a3"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "quick-xml"
version = "0.32.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1d3a6e5838b60e0e8fa7a43f22ade549a37d61f8bdbe636d0d7816191de969c2"
dependencies = [
"memchr",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "quote"
version = "1.0.40"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1885c039570dc00dcb4ff087a89e185fd56bae234ddc7f056a945bf36467248d"
dependencies = [
"proc-macro2",
]
[[package]]
name = "r-efi"
version = "5.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "74765f6d916ee2faa39bc8e68e4f3ed8949b48cccdac59983d287a7cb71ce9c5"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "radix_trie"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c069c179fcdc6a2fe24d8d18305cf085fdbd4f922c041943e203685d6a1c58fd"
dependencies = [
"endian-type",
"nibble_vec",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "rand"
version = "0.8.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "34af8d1a0e25924bc5b7c43c079c942339d8f0a8b57c39049bef581b46327404"
dependencies = [
"libc",
"rand_chacha 0.3.1",
"rand_core 0.6.4",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "rand"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9fbfd9d094a40bf3ae768db9361049ace4c0e04a4fd6b359518bd7b73a73dd97"
dependencies = [
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"rand_chacha 0.9.0",
"rand_core 0.9.3",
]
[[package]]
name = "rand_chacha"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e6c10a63a0fa32252be49d21e7709d4d4baf8d231c2dbce1eaa8141b9b127d88"
dependencies = [
"ppv-lite86",
"rand_core 0.6.4",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "rand_chacha"
version = "0.9.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d3022b5f1df60f26e1ffddd6c66e8aa15de382ae63b3a0c1bfc0e4d3e3f325cb"
dependencies = [
"ppv-lite86",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"rand_core 0.9.3",
]
[[package]]
name = "rand_core"
version = "0.6.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c"
dependencies = [
"getrandom 0.2.16",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "rand_core"
version = "0.9.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "99d9a13982dcf210057a8a78572b2217b667c3beacbf3a0d8b454f6f82837d38"
dependencies = [
"getrandom 0.3.2",
]
[[package]]
name = "ratatui"
version = "0.29.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "eabd94c2f37801c20583fc49dd5cd6b0ba68c716787c2dd6ed18571e1e63117b"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"cassowary",
"compact_str",
"crossterm",
"indoc",
"instability",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"itertools 0.13.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"lru",
"paste",
feat: add support for commands in the Rust TUI (#935) Introduces support for slash commands like in the TypeScript CLI. We do not support the full set of commands yet, but the core abstraction is there now. In particular, we have a `SlashCommand` enum and due to thoughtful use of the [strum](https://crates.io/crates/strum) crate, it requires minimal boilerplate to add a new command to the list. The key new piece of UI is `CommandPopup`, though the keyboard events are still handled by `ChatComposer`. The behavior is roughly as follows: * if the first character in the composer is `/`, the command popup is displayed (if you really want to send a message to Codex that starts with a `/`, simply put a space before the `/`) * while the popup is displayed, up/down can be used to change the selection of the popup * if there is a selection, hitting tab completes the command, but does not send it * if there is a selection, hitting enter sends the command * if the prefix of the composer matches a command, the command will be visible in the popup so the user can see the description (commands could take arguments, so additional text may appear after the command name itself) https://github.com/user-attachments/assets/39c3e6ee-eeb7-4ef7-a911-466d8184975f Incidentally, Codex wrote almost all the code for this PR!
2025-05-14 12:55:49 -07:00
"strum 0.26.3",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"unicode-segmentation",
"unicode-truncate",
"unicode-width 0.2.0",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "ratatui-image"
version = "8.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d3f1d31464920104b247593f008158372d2fdb8165e93a4299cdd6f994448c9a"
dependencies = [
"base64 0.21.7",
"icy_sixel",
"image",
"rand 0.8.5",
"ratatui",
"rustix 0.38.44",
"thiserror 1.0.69",
"windows",
]
[[package]]
name = "rav1e"
version = "0.7.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cd87ce80a7665b1cce111f8a16c1f3929f6547ce91ade6addf4ec86a8dda5ce9"
dependencies = [
"arbitrary",
"arg_enum_proc_macro",
"arrayvec",
"av1-grain",
"bitstream-io",
"built",
"cfg-if",
"interpolate_name",
"itertools 0.12.1",
"libc",
"libfuzzer-sys",
"log",
"maybe-rayon",
"new_debug_unreachable",
"noop_proc_macro",
"num-derive",
"num-traits",
"once_cell",
"paste",
"profiling",
"rand 0.8.5",
"rand_chacha 0.3.1",
"simd_helpers",
"system-deps",
"thiserror 1.0.69",
"v_frame",
"wasm-bindgen",
]
[[package]]
name = "ravif"
version = "0.11.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d6a5f31fcf7500f9401fea858ea4ab5525c99f2322cfcee732c0e6c74208c0c6"
dependencies = [
"avif-serialize",
"imgref",
"loop9",
"quick-error",
"rav1e",
"rayon",
"rgb",
]
[[package]]
name = "rayon"
version = "1.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b418a60154510ca1a002a752ca9714984e21e4241e804d32555251faf8b78ffa"
dependencies = [
"either",
"rayon-core",
]
[[package]]
name = "rayon-core"
version = "1.12.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1465873a3dfdaa8ae7cb14b4383657caab0b3e8a0aa9ae8e04b044854c8dfce2"
dependencies = [
"crossbeam-deque",
"crossbeam-utils",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "redox_syscall"
version = "0.5.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d2f103c6d277498fbceb16e84d317e2a400f160f46904d5f5410848c829511a3"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "redox_users"
version = "0.4.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ba009ff324d1fc1b900bd1fdb31564febe58a8ccc8a6fdbb93b543d33b13ca43"
dependencies = [
"getrandom 0.2.16",
"libredox",
"thiserror 1.0.69",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "redox_users"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dd6f9d3d47bdd2ad6945c5015a226ec6155d0bcdfd8f7cd29f86b71f8de99d2b"
dependencies = [
"getrandom 0.2.16",
"libredox",
"thiserror 2.0.12",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "ref-cast"
version = "1.0.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4a0ae411dbe946a674d89546582cea4ba2bb8defac896622d6496f14c23ba5cf"
dependencies = [
"ref-cast-impl",
]
[[package]]
name = "ref-cast-impl"
version = "1.0.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1165225c21bff1f3bbce98f5a1f889949bc902d3575308cc7b0de30b4f6d27c7"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "regex"
version = "1.11.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b544ef1b4eac5dc2db33ea63606ae9ffcfac26c1416a2806ae0bf5f56b201191"
dependencies = [
"aho-corasick",
"memchr",
"regex-automata 0.4.9",
"regex-syntax 0.8.5",
]
[[package]]
name = "regex-automata"
version = "0.1.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6c230d73fb8d8c1b9c0b3135c5142a8acee3a0558fb8db5cf1cb65f8d7862132"
dependencies = [
"regex-syntax 0.6.29",
]
[[package]]
name = "regex-automata"
version = "0.4.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "809e8dc61f6de73b46c85f4c96486310fe304c434cfa43669d7b40f711150908"
dependencies = [
"aho-corasick",
"memchr",
"regex-syntax 0.8.5",
]
chore: replace regex with regex-lite, where appropriate (#1200) As explained on https://crates.io/crates/regex-lite, `regex-lite` is a lighter alternative to `regex` and seems to be sufficient for our purposes.
2025-06-02 17:11:45 -07:00
[[package]]
name = "regex-lite"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "53a49587ad06b26609c52e423de037e7f57f20d53535d66e08c695f347df952a"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "regex-syntax"
version = "0.6.29"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f162c6dd7b008981e4d40210aca20b4bd0f9b60ca9271061b07f78537722f2e1"
[[package]]
name = "regex-syntax"
version = "0.8.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2b15c43186be67a4fd63bee50d0303afffcef381492ebe2c5d87f324e1b8815c"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "relative-path"
version = "1.9.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ba39f3699c378cd8970968dcbff9c43159ea4cfbd88d43c00b22f2ef10a435d2"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "reqwest"
version = "0.12.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d19c46a6fdd48bc4dab94b6103fccc55d34c67cc0ad04653aad4ea2a07cd7bbb"
dependencies = [
"base64 0.22.1",
"bytes",
"encoding_rs",
"futures-core",
"futures-util",
"h2",
"http",
"http-body",
"http-body-util",
"hyper",
"hyper-rustls",
"hyper-tls",
"hyper-util",
"ipnet",
"js-sys",
"log",
"mime",
"native-tls",
"once_cell",
"percent-encoding",
"pin-project-lite",
"rustls-pemfile",
"serde",
"serde_json",
"serde_urlencoded",
"sync_wrapper",
"system-configuration",
"tokio",
"tokio-native-tls",
"tokio-util",
"tower",
"tower-service",
"url",
"wasm-bindgen",
"wasm-bindgen-futures",
"wasm-streams",
"web-sys",
"windows-registry",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "rgb"
version = "0.8.50"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "57397d16646700483b67d2dd6511d79318f9d057fdbd21a4066aeac8b41d310a"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "ring"
version = "0.17.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a4689e6c2294d81e88dc6261c768b63bc4fcdb852be6d1352498b114f61383b7"
dependencies = [
"cc",
"cfg-if",
"getrandom 0.2.16",
"libc",
"untrusted",
"windows-sys 0.52.0",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "rstest"
version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6fc39292f8613e913f7df8fa892b8944ceb47c247b78e1b1ae2f09e019be789d"
dependencies = [
"futures-timer",
"futures-util",
"rstest_macros",
"rustc_version",
]
[[package]]
name = "rstest_macros"
version = "0.25.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1f168d99749d307be9de54d23fd226628d99768225ef08f6ffb52e0182a27746"
dependencies = [
"cfg-if",
"glob",
"proc-macro-crate",
"proc-macro2",
"quote",
"regex",
"relative-path",
"rustc_version",
"syn 2.0.100",
"unicode-ident",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "rustc-demangle"
version = "0.1.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "719b953e2095829ee67db738b3bfa9fa368c94900df327b3f07fe6e794d2fe1f"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "rustc_version"
version = "0.4.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cfcb3a22ef46e85b45de6ee7e79d063319ebb6594faafcf1c225ea92ab6e9b92"
dependencies = [
"semver",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "rustix"
version = "0.38.44"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fdb5bc1ae2baa591800df16c9ca78619bf65c0488b41b96ccec5d11220d8c154"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"errno",
"libc",
"linux-raw-sys 0.4.15",
"windows-sys 0.59.0",
]
[[package]]
name = "rustix"
version = "1.0.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d97817398dd4bb2e6da002002db259209759911da105da92bec29ccb12cf58bf"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"errno",
"libc",
"linux-raw-sys 0.9.4",
"windows-sys 0.59.0",
]
[[package]]
name = "rustls"
version = "0.23.26"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "df51b5869f3a441595eac5e8ff14d486ff285f7b8c0df8770e49c3b56351f0f0"
dependencies = [
"once_cell",
"rustls-pki-types",
"rustls-webpki",
"subtle",
"zeroize",
]
[[package]]
name = "rustls-pemfile"
version = "2.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dce314e5fee3f39953d46bb63bb8a46d40c2f8fb7cc5a3b6cab2bde9721d6e50"
dependencies = [
"rustls-pki-types",
]
[[package]]
name = "rustls-pki-types"
version = "1.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "917ce264624a4b4db1c364dcc35bfca9ded014d0a958cd47ad3e960e988ea51c"
[[package]]
name = "rustls-webpki"
version = "0.103.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fef8b8769aaccf73098557a87cd1816b4f9c7c16811c9c77142aa695c16f2c03"
dependencies = [
"ring",
"rustls-pki-types",
"untrusted",
]
[[package]]
name = "rustversion"
version = "1.0.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "eded382c5f5f786b989652c49544c4877d9f015cc22e145a5ea8ea66c2921cd2"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "rustyline"
version = "14.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7803e8936da37efd9b6d4478277f4b2b9bb5cdb37a113e8d63222e58da647e63"
dependencies = [
"bitflags 2.9.0",
"cfg-if",
"clipboard-win",
"fd-lock",
"home",
"libc",
"log",
"memchr",
"nix",
"radix_trie",
"unicode-segmentation",
"unicode-width 0.1.14",
"utf8parse",
"windows-sys 0.52.0",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "ryu"
version = "1.0.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "28d3b2b1366ec20994f1fd18c3c594f05c5dd4bc44d8bb0c1c632c8d6829481f"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "same-file"
version = "1.0.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "93fc1dc3aaa9bfed95e02e6eadabb4baf7e3078b0bd1b4d7b6b0b68378900502"
dependencies = [
"winapi-util",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "schannel"
version = "0.1.27"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1f29ebaa345f945cec9fbbc532eb307f0fdad8161f281b6369539c8d84876b3d"
dependencies = [
"windows-sys 0.59.0",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "schemafy"
version = "0.5.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8aea5ba40287dae331f2c48b64dbc8138541f5e97ee8793caa7948c1f31d86d5"
dependencies = [
"Inflector",
"schemafy_core",
"schemafy_lib",
"serde",
"serde_derive",
"serde_json",
"serde_repr",
"syn 1.0.109",
]
[[package]]
name = "schemafy_core"
version = "0.5.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "41781ae092f4fd52c9287efb74456aea0d3b90032d2ecad272bd14dbbcb0511b"
dependencies = [
"serde",
"serde_json",
]
[[package]]
name = "schemafy_lib"
version = "0.5.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e953db32579999ca98c451d80801b6f6a7ecba6127196c5387ec0774c528befa"
dependencies = [
"Inflector",
"proc-macro2",
"quote",
"schemafy_core",
"serde",
"serde_derive",
"serde_json",
"syn 1.0.109",
]
feat: make Codex available as a tool when running it as an MCP server (#811) This PR replaces the placeholder `"echo"` tool call in the MCP server with a `"codex"` tool that calls Codex. Events such as `ExecApprovalRequest` and `ApplyPatchApprovalRequest` are not handled properly yet, but I have `approval_policy = "never"` set in my `~/.codex/config.toml` such that those codepaths are not exercised. The schema for this MPC tool is defined by a new `CodexToolCallParam` struct introduced in this PR. It is fairly similar to `ConfigOverrides`, as the param is used to help create the `Config` used to start the Codex session, though it also includes the `prompt` used to kick off the session. This PR also introduces the use of the third-party `schemars` crate to generate the JSON schema, which is verified in the `verify_codex_tool_json_schema()` unit test. Events that are dispatched during the Codex session are sent back to the MCP client as MCP notifications. This gives the client a way to monitor progress as the tool call itself may take minutes to complete depending on the complexity of the task requested by the user. In the video below, I launched the server via: ```shell mcp-server$ RUST_LOG=debug npx @modelcontextprotocol/inspector cargo run -- ``` In the video, you can see the flow of: * requesting the list of tools * choosing the **codex** tool * entering a value for **prompt** and then making the tool call Note that I left the other fields blank because when unspecified, the values in my `~/.codex/config.toml` were used: https://github.com/user-attachments/assets/1975058c-b004-43ef-8c8d-800a953b8192 Note that while using the inspector, I did run into https://github.com/modelcontextprotocol/inspector/issues/293, though the tip about ensuring I had only one instance of the **MCP Inspector** tab open in my browser seemed to fix things.
2025-05-05 07:16:19 -07:00
[[package]]
name = "schemars"
version = "0.8.22"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3fbf2ae1b8bc8e02df939598064d22402220cd5bbcca1c76f7d6a310974d5615"
dependencies = [
"dyn-clone",
"schemars_derive",
"serde",
"serde_json",
]
[[package]]
name = "schemars_derive"
version = "0.8.22"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "32e265784ad618884abaea0600a9adf15393368d840e0222d101a072f3f7534d"
dependencies = [
"proc-macro2",
"quote",
"serde_derive_internals",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "scopeguard"
version = "1.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
[[package]]
name = "seccompiler"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a4ae55de56877481d112a559bbc12667635fdaf5e005712fd4e2b2fa50ffc884"
dependencies = [
"libc",
]
[[package]]
name = "security-framework"
version = "2.11.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "897b2245f0b511c87893af39b033e5ca9cce68824c4d7e7630b5a1d339658d02"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"core-foundation",
"core-foundation-sys",
"libc",
"security-framework-sys",
]
[[package]]
name = "security-framework-sys"
version = "2.14.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "49db231d56a190491cb4aeda9527f1ad45345af50b0851622a7adb8c03b01c32"
dependencies = [
"core-foundation-sys",
"libc",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "semver"
version = "1.0.26"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56e6fa9c48d24d85fb3de5ad847117517440f6beceb7798af16b4a87d616b8d0"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "serde"
version = "1.0.219"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5f0e2c6ed6606019b4e29e69dbaba95b11854410e5347d525002456dbbb786b6"
dependencies = [
"serde_derive",
]
[[package]]
name = "serde_derive"
version = "1.0.219"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5b0276cf7f2c73365f7157c8123c21cd9a50fbbd844757af28ca1f5925fc2a00"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: make Codex available as a tool when running it as an MCP server (#811) This PR replaces the placeholder `"echo"` tool call in the MCP server with a `"codex"` tool that calls Codex. Events such as `ExecApprovalRequest` and `ApplyPatchApprovalRequest` are not handled properly yet, but I have `approval_policy = "never"` set in my `~/.codex/config.toml` such that those codepaths are not exercised. The schema for this MPC tool is defined by a new `CodexToolCallParam` struct introduced in this PR. It is fairly similar to `ConfigOverrides`, as the param is used to help create the `Config` used to start the Codex session, though it also includes the `prompt` used to kick off the session. This PR also introduces the use of the third-party `schemars` crate to generate the JSON schema, which is verified in the `verify_codex_tool_json_schema()` unit test. Events that are dispatched during the Codex session are sent back to the MCP client as MCP notifications. This gives the client a way to monitor progress as the tool call itself may take minutes to complete depending on the complexity of the task requested by the user. In the video below, I launched the server via: ```shell mcp-server$ RUST_LOG=debug npx @modelcontextprotocol/inspector cargo run -- ``` In the video, you can see the flow of: * requesting the list of tools * choosing the **codex** tool * entering a value for **prompt** and then making the tool call Note that I left the other fields blank because when unspecified, the values in my `~/.codex/config.toml` were used: https://github.com/user-attachments/assets/1975058c-b004-43ef-8c8d-800a953b8192 Note that while using the inspector, I did run into https://github.com/modelcontextprotocol/inspector/issues/293, though the tip about ensuring I had only one instance of the **MCP Inspector** tab open in my browser seemed to fix things.
2025-05-05 07:16:19 -07:00
[[package]]
name = "serde_derive_internals"
version = "0.29.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "18d26a20a969b9e3fdf2fc2d9f21eda6c40e2de84c9408bb5d3b05d499aae711"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "serde_json"
version = "1.0.140"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "20068b6e96dc6c9bd23e01df8827e6c7e1f2fddd43c21810382803c136b99373"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"indexmap 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"itoa",
"memchr",
"ryu",
"serde",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "serde_repr"
version = "0.1.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "175ee3e80ae9982737ca543e96133087cbd9a485eecc3bc4de9c1a37b47ea59c"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "serde_spanned"
version = "0.6.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "87607cb1398ed59d48732e575a4c28a7a8ebf2454b964fe3f224f2afc07909e1"
dependencies = [
"serde",
]
[[package]]
name = "serde_urlencoded"
version = "0.7.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d3491c14715ca2294c4d6a88f15e84739788c1d030eed8c110436aafdaa2f3fd"
dependencies = [
"form_urlencoded",
"itoa",
"ryu",
"serde",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "serde_with"
version = "3.12.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d6b6f7f2fcb69f747921f79f3926bd1e203fce4fef62c268dd3abfb6d86029aa"
dependencies = [
"base64 0.22.1",
"chrono",
"hex",
"indexmap 1.9.3",
"indexmap 2.9.0",
"serde",
"serde_derive",
"serde_json",
"serde_with_macros",
"time",
]
[[package]]
name = "serde_with_macros"
version = "3.12.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8d00caa5193a3c8362ac2b73be6b9e768aa5a4b2f721d8f4b339600c3cb51f8e"
dependencies = [
"darling",
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "sharded-slab"
version = "0.1.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f40ca3c46823713e0d4209592e8d6e826aa57e928f09752619fc696c499637f6"
dependencies = [
"lazy_static",
]
[[package]]
name = "shlex"
version = "1.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0fda2ff0d084019ba4d7c6f371c95d8fd75ce3524c3cb8fb653a3023f6323e64"
[[package]]
name = "signal-hook"
version = "0.3.17"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8621587d4798caf8eb44879d42e56b9a93ea5dcd315a6487c357130095b62801"
dependencies = [
"libc",
"signal-hook-registry",
]
[[package]]
name = "signal-hook-mio"
version = "0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "34db1a06d485c9142248b7a054f034b349b212551f3dfd19c94d45a754a217cd"
dependencies = [
"libc",
"mio",
"signal-hook",
]
[[package]]
name = "signal-hook-registry"
version = "1.4.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9203b8055f63a2a00e2f593bb0510367fe707d7ff1e5c872de2f537b339e5410"
dependencies = [
"libc",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "simd-adler32"
version = "0.3.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d66dc143e6b11c1eddc06d5c423cfc97062865baf299914ab64caa38182078fe"
[[package]]
name = "simd_helpers"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "95890f873bec569a0362c235787f3aca6e1e887302ba4840839bcc6459c42da6"
dependencies = [
"quote",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "simdutf8"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e3a9fe34e3e7a50316060351f37187a3f546bce95496156754b601a5fa71b76e"
[[package]]
name = "similar"
version = "2.7.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bbbb5d9659141646ae647b42fe094daf6c6192d1620870b449d9557f748b2daa"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "siphasher"
version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56199f7ddabf13fe5074ce809e7d3f42b42ae711800501b5b16ea82ad029c39d"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "slab"
version = "0.4.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8f92a496fb766b417c996b9c5e57daf2f7ad3b0bebe1ccfca4856390e3d3bb67"
dependencies = [
"autocfg",
]
[[package]]
name = "smallvec"
version = "1.15.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8917285742e9f3e1683f0a9c4e6b57960b7314d0b08d30d1ecd426713ee2eee9"
[[package]]
name = "socket2"
version = "0.5.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4f5fd57c80058a56cf5c777ab8a126398ece8e442983605d280a44ce79d0edef"
dependencies = [
"libc",
"windows-sys 0.52.0",
]
[[package]]
name = "stable_deref_trait"
version = "1.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "starlark"
version = "0.13.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0f53849859f05d9db705b221bd92eede93877fd426c1b4a3c3061403a5912a8f"
dependencies = [
"allocative",
"anyhow",
"bumpalo",
"cmp_any",
"debugserver-types",
"derivative",
"derive_more",
"display_container",
"dupe",
"either",
"erased-serde",
"hashbrown 0.14.5",
"inventory",
"itertools 0.13.0",
"maplit",
"memoffset",
"num-bigint",
"num-traits",
"once_cell",
"paste",
"ref-cast",
"regex",
"rustyline",
"serde",
"serde_json",
"starlark_derive",
"starlark_map",
"starlark_syntax",
"static_assertions",
"strsim 0.10.0",
"textwrap",
"thiserror 1.0.69",
]
[[package]]
name = "starlark_derive"
version = "0.13.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fe58bc6c8b7980a1fe4c9f8f48200c3212db42ebfe21ae6a0336385ab53f082a"
dependencies = [
"dupe",
"proc-macro2",
"quote",
"syn 2.0.100",
]
[[package]]
name = "starlark_map"
version = "0.13.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "92659970f120df0cc1c0bb220b33587b7a9a90e80d4eecc5c5af5debb950173d"
dependencies = [
"allocative",
"dupe",
"equivalent",
"fxhash",
"hashbrown 0.14.5",
"serde",
]
[[package]]
name = "starlark_syntax"
version = "0.13.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fe53b3690d776aafd7cb6b9fed62d94f83280e3b87d88e3719cc0024638461b3"
dependencies = [
"allocative",
"annotate-snippets",
"anyhow",
"derivative",
"derive_more",
"dupe",
"lalrpop",
"lalrpop-util",
"logos",
"lsp-types",
"memchr",
"num-bigint",
"num-traits",
"once_cell",
"starlark_map",
"thiserror 1.0.69",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "static_assertions"
version = "1.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f"
[[package]]
name = "streaming-iterator"
version = "0.1.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2b2231b7c3057d5e4ad0156fb3dc807d900806020c5ffa3ee6ff2c8c76fb8520"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "string_cache"
version = "0.8.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bf776ba3fa74f83bf4b63c3dcbbf82173db2632ed8452cb2d891d33f459de70f"
dependencies = [
"new_debug_unreachable",
"parking_lot",
"phf_shared",
"precomputed-hash",
]
[[package]]
name = "strsim"
version = "0.10.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "73473c0e59e6d5812c5dfe2a064a6444949f089e20eec9a2e5506596494e4623"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "strsim"
version = "0.11.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7da8b5736845d9f2fcb837ea5d9e2628564b3b043a70948a3f0b778838c5fb4f"
[[package]]
name = "strum"
version = "0.26.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8fec0f0aef304996cf250b31b5a10dee7980c85da9d759361292b8bca5a18f06"
dependencies = [
feat: add support for commands in the Rust TUI (#935) Introduces support for slash commands like in the TypeScript CLI. We do not support the full set of commands yet, but the core abstraction is there now. In particular, we have a `SlashCommand` enum and due to thoughtful use of the [strum](https://crates.io/crates/strum) crate, it requires minimal boilerplate to add a new command to the list. The key new piece of UI is `CommandPopup`, though the keyboard events are still handled by `ChatComposer`. The behavior is roughly as follows: * if the first character in the composer is `/`, the command popup is displayed (if you really want to send a message to Codex that starts with a `/`, simply put a space before the `/`) * while the popup is displayed, up/down can be used to change the selection of the popup * if there is a selection, hitting tab completes the command, but does not send it * if there is a selection, hitting enter sends the command * if the prefix of the composer matches a command, the command will be visible in the popup so the user can see the description (commands could take arguments, so additional text may appear after the command name itself) https://github.com/user-attachments/assets/39c3e6ee-eeb7-4ef7-a911-466d8184975f Incidentally, Codex wrote almost all the code for this PR!
2025-05-14 12:55:49 -07:00
"strum_macros 0.26.4",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: add support for commands in the Rust TUI (#935) Introduces support for slash commands like in the TypeScript CLI. We do not support the full set of commands yet, but the core abstraction is there now. In particular, we have a `SlashCommand` enum and due to thoughtful use of the [strum](https://crates.io/crates/strum) crate, it requires minimal boilerplate to add a new command to the list. The key new piece of UI is `CommandPopup`, though the keyboard events are still handled by `ChatComposer`. The behavior is roughly as follows: * if the first character in the composer is `/`, the command popup is displayed (if you really want to send a message to Codex that starts with a `/`, simply put a space before the `/`) * while the popup is displayed, up/down can be used to change the selection of the popup * if there is a selection, hitting tab completes the command, but does not send it * if there is a selection, hitting enter sends the command * if the prefix of the composer matches a command, the command will be visible in the popup so the user can see the description (commands could take arguments, so additional text may appear after the command name itself) https://github.com/user-attachments/assets/39c3e6ee-eeb7-4ef7-a911-466d8184975f Incidentally, Codex wrote almost all the code for this PR!
2025-05-14 12:55:49 -07:00
[[package]]
name = "strum"
version = "0.27.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f64def088c51c9510a8579e3c5d67c65349dcf755e5479ad3d010aa6454e2c32"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "strum_macros"
version = "0.26.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4c6bee85a5a24955dc440386795aa378cd9cf82acd5f764469152d2270e581be"
dependencies = [
"heck",
"proc-macro2",
"quote",
"rustversion",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: add support for commands in the Rust TUI (#935) Introduces support for slash commands like in the TypeScript CLI. We do not support the full set of commands yet, but the core abstraction is there now. In particular, we have a `SlashCommand` enum and due to thoughtful use of the [strum](https://crates.io/crates/strum) crate, it requires minimal boilerplate to add a new command to the list. The key new piece of UI is `CommandPopup`, though the keyboard events are still handled by `ChatComposer`. The behavior is roughly as follows: * if the first character in the composer is `/`, the command popup is displayed (if you really want to send a message to Codex that starts with a `/`, simply put a space before the `/`) * while the popup is displayed, up/down can be used to change the selection of the popup * if there is a selection, hitting tab completes the command, but does not send it * if there is a selection, hitting enter sends the command * if the prefix of the composer matches a command, the command will be visible in the popup so the user can see the description (commands could take arguments, so additional text may appear after the command name itself) https://github.com/user-attachments/assets/39c3e6ee-eeb7-4ef7-a911-466d8184975f Incidentally, Codex wrote almost all the code for this PR!
2025-05-14 12:55:49 -07:00
[[package]]
name = "strum_macros"
version = "0.27.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c77a8c5abcaf0f9ce05d62342b7d298c346515365c36b673df4ebe3ced01fde8"
dependencies = [
"heck",
"proc-macro2",
"quote",
"rustversion",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "subtle"
version = "2.6.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "13c2bddecc57b384dee18652358fb23172facb8a2c51ccc10d74c157bdea3292"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "syn"
version = "1.0.109"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "72b64191b275b66ffe2469e8af2c1cfe3bafa67b529ead792a6d0160888b4237"
dependencies = [
"proc-macro2",
"quote",
"unicode-ident",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "syn"
version = "2.0.100"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b09a44accad81e1ba1cd74a32461ba89dee89095ba17b32f5d03683b1b1fc2a0"
dependencies = [
"proc-macro2",
"quote",
"unicode-ident",
]
[[package]]
name = "sync_wrapper"
version = "1.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0bf256ce5efdfa370213c1dabab5935a12e49f2c58d15e9eac2870d3b4f27263"
dependencies = [
"futures-core",
]
[[package]]
name = "synstructure"
version = "0.13.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c8af7666ab7b6390ab78131fb5b0fce11d6b7a6951602017c35fa82800708971"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "syntect"
version = "5.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "874dcfa363995604333cf947ae9f751ca3af4522c60886774c4963943b4746b1"
dependencies = [
"bincode",
"bitflags 1.3.2",
"flate2",
"fnv",
"once_cell",
"onig",
"plist",
"regex-syntax 0.8.5",
"serde",
"serde_derive",
"serde_json",
"thiserror 1.0.69",
"walkdir",
"yaml-rust",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "system-configuration"
version = "0.6.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3c879d448e9d986b661742763247d3693ed13609438cf3d006f51f5368a5ba6b"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"core-foundation",
"system-configuration-sys",
]
[[package]]
name = "system-configuration-sys"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8e1d1b10ced5ca923a1fcb8d03e96b8d3268065d724548c0211415ff6ac6bac4"
dependencies = [
"core-foundation-sys",
"libc",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "system-deps"
version = "6.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a3e535eb8dded36d55ec13eddacd30dec501792ff23a0b1682c38601b8cf2349"
dependencies = [
"cfg-expr",
"heck",
"pkg-config",
"toml",
"version-compare",
]
[[package]]
name = "target-lexicon"
version = "0.12.16"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "61c41af27dd6d1e27b1b16b489db798443478cef1f06a660c96db617ba5de3b1"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "tempfile"
version = "3.19.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7437ac7763b9b123ccf33c338a5cc1bac6f69b45a136c19bdd8a65e3916435bf"
dependencies = [
"fastrand",
"getrandom 0.3.2",
"once_cell",
"rustix 1.0.5",
"windows-sys 0.59.0",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "term"
version = "0.7.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c59df8ac95d96ff9bede18eb7300b0fda5e5d8d90960e76f8e14ae765eedbf1f"
dependencies = [
"dirs-next",
"rustversion",
"winapi",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "terminal_size"
version = "0.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "45c6481c4829e4cc63825e62c49186a34538b7b2750b73b266581ffb612fb5ed"
dependencies = [
"rustix 1.0.5",
"windows-sys 0.59.0",
]
[[package]]
name = "termtree"
version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8f50febec83f5ee1df3015341d8bd429f2d1cc62bcba7ea2076759d315084683"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "textwrap"
version = "0.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d326610f408c7a4eb6f51c37c330e496b08506c9457c9d34287ecc38809fb060"
dependencies = [
"unicode-width 0.1.14",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "thiserror"
version = "1.0.69"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b6aaf5339b578ea85b50e080feb250a3e8ae8cfcdff9a461c9ec2904bc923f52"
dependencies = [
"thiserror-impl 1.0.69",
]
[[package]]
name = "thiserror"
version = "2.0.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "567b8a2dae586314f7be2a752ec7474332959c6460e02bde30d702a66d488708"
dependencies = [
"thiserror-impl 2.0.12",
]
[[package]]
name = "thiserror-impl"
version = "1.0.69"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4fee6c4efc90059e10f81e6d42c60a18f76588c3d74cb83a0b242a2b6c7504c1"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "thiserror-impl"
version = "2.0.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7f7cf42b4507d8ea322120659672cf1b9dbb93f8f2d4ecfd6e51350ff5b17a1d"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "thread_local"
version = "1.1.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8b9ef9bad013ada3808854ceac7b46812a6465ba368859a37e2100283d2d719c"
dependencies = [
"cfg-if",
"once_cell",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "tiff"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ba1310fcea54c6a9a4fd1aad794ecc02c31682f6bfbecdf460bf19533eed1e3e"
dependencies = [
"flate2",
"jpeg-decoder",
"weezl",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "time"
version = "0.3.41"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8a7619e19bc266e0f9c5e6686659d394bc57973859340060a69221e57dbc0c40"
dependencies = [
"deranged",
"itoa",
fix: tighten up some logic around session timestamps and ids (#922) * update `SessionConfigured` event to include the UUID for the session * show the UUID in the Rust TUI * use local timestamps in log files instead of UTC * include timestamps in log file names for easier discovery
2025-05-13 19:22:16 -07:00
"libc",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"num-conv",
fix: tighten up some logic around session timestamps and ids (#922) * update `SessionConfigured` event to include the UUID for the session * show the UUID in the Rust TUI * use local timestamps in log files instead of UTC * include timestamps in log file names for easier discovery
2025-05-13 19:22:16 -07:00
"num_threads",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"powerfmt",
"serde",
"time-core",
"time-macros",
]
[[package]]
name = "time-core"
version = "0.1.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c9e9a38711f559d9e3ce1cdb06dd7c5b8ea546bc90052da6d06bb76da74bb07c"
[[package]]
name = "time-macros"
version = "0.2.22"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3526739392ec93fd8b359c8e98514cb3e8e021beb4e5f597b00a0221f8ed8a49"
dependencies = [
"num-conv",
"time-core",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "tiny-keccak"
version = "2.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2c9d3793400a45f954c52e73d068316d76b6f4e36977e3fcebb13a2721e80237"
dependencies = [
"crunchy",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "tinystr"
version = "0.7.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9117f5d4db391c1cf6927e7bea3db74b9a1c1add8f7eda9ffd5364f40f57b82f"
dependencies = [
"displaydoc",
"zerovec",
]
[[package]]
name = "tokio"
version = "1.44.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e6b88822cbe49de4185e3a4cbf8321dd487cf5fe0c5c65695fef6346371e9c48"
dependencies = [
"backtrace",
"bytes",
"libc",
"mio",
"pin-project-lite",
"signal-hook-registry",
"socket2",
"tokio-macros",
"windows-sys 0.52.0",
]
[[package]]
name = "tokio-macros"
version = "2.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6e06d43f1345a3bcd39f6a56dbb7dcab2ba47e68e8ac134855e7e2bdbaf8cab8"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "tokio-native-tls"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bbae76ab933c85776efabc971569dd6119c580d8f5d448769dec1764bf796ef2"
dependencies = [
"native-tls",
"tokio",
]
[[package]]
name = "tokio-rustls"
version = "0.26.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8e727b36a1a0e8b74c376ac2211e40c2c8af09fb4013c60d910495810f008e9b"
dependencies = [
"rustls",
"tokio",
]
[[package]]
name = "tokio-util"
version = "0.7.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6b9590b93e6fcc1739458317cccd391ad3955e2bde8913edf6f95f9e65a8f034"
dependencies = [
"bytes",
"futures-core",
"futures-sink",
"pin-project-lite",
"tokio",
]
[[package]]
name = "toml"
version = "0.8.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cd87a5cdd6ffab733b2f74bc4fd7ee5fff6634124999ac278c35fc78c6120148"
dependencies = [
"serde",
"serde_spanned",
"toml_datetime",
"toml_edit",
]
[[package]]
name = "toml_datetime"
version = "0.6.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0dd7358ecb8fc2f8d014bf86f6f638ce72ba252a2c3a2572f2a795f1d23efb41"
dependencies = [
"serde",
]
[[package]]
name = "toml_edit"
version = "0.22.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "17b4795ff5edd201c7cd6dca065ae59972ce77d1b80fa0a84d94950ece7d1474"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"indexmap 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"serde",
"serde_spanned",
"toml_datetime",
"winnow",
]
[[package]]
name = "tower"
version = "0.5.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d039ad9159c98b70ecfd540b2573b97f7f52c3e8d9f8ad57a24b916a536975f9"
dependencies = [
"futures-core",
"futures-util",
"pin-project-lite",
"sync_wrapper",
"tokio",
"tower-layer",
"tower-service",
]
[[package]]
name = "tower-layer"
version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "121c2a6cda46980bb0fcd1647ffaf6cd3fc79a013de288782836f6df9c48780e"
[[package]]
name = "tower-service"
version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8df9b6e13f2d32c91b9bd719c00d1958837bc7dec474d94952798cc8e69eeec3"
[[package]]
name = "tracing"
version = "0.1.41"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "784e0ac535deb450455cbfa28a6f0df145ea1bb7ae51b821cf5e7927fdcfbdd0"
dependencies = [
"log",
"pin-project-lite",
"tracing-attributes",
"tracing-core",
]
[[package]]
name = "tracing-appender"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3566e8ce28cc0a3fe42519fc80e6b4c943cc4c8cef275620eb8dac2d3d4e06cf"
dependencies = [
"crossbeam-channel",
"thiserror 1.0.69",
"time",
"tracing-subscriber",
]
[[package]]
name = "tracing-attributes"
version = "0.1.28"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "395ae124c09f9e6918a2310af6038fba074bcf474ac352496d5910dd59a2226d"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "tracing-core"
version = "0.1.33"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e672c95779cf947c5311f83787af4fa8fffd12fb27e4993211a84bdfd9610f9c"
dependencies = [
"once_cell",
"valuable",
]
[[package]]
name = "tracing-error"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8b1581020d7a273442f5b45074a6a57d5757ad0a47dac0e9f0bd57b81936f3db"
dependencies = [
"tracing",
"tracing-subscriber",
]
[[package]]
name = "tracing-log"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ee855f1f400bd0e5c02d150ae5de3840039a3f54b025156404e34c23c03f47c3"
dependencies = [
"log",
"once_cell",
"tracing-core",
]
[[package]]
name = "tracing-subscriber"
version = "0.3.19"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e8189decb5ac0fa7bc8b96b7cb9b2701d60d48805aca84a238004d665fcc4008"
dependencies = [
"matchers",
"nu-ansi-term",
"once_cell",
"regex",
"sharded-slab",
"smallvec",
"thread_local",
"tracing",
"tracing-core",
"tracing-log",
]
[[package]]
name = "tree-sitter"
version = "0.25.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b9ac5ea5e7f2f1700842ec071401010b9c59bf735295f6e9fa079c3dc035b167"
dependencies = [
"cc",
"regex",
"regex-syntax 0.8.5",
"serde_json",
"streaming-iterator",
"tree-sitter-language",
]
[[package]]
name = "tree-sitter-bash"
version = "0.23.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "329a4d48623ac337d42b1df84e81a1c9dbb2946907c102ca72db158c1964a52e"
dependencies = [
"cc",
"tree-sitter-language",
]
[[package]]
name = "tree-sitter-language"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c4013970217383f67b18aef68f6fb2e8d409bc5755227092d32efb0422ba24b8"
[[package]]
name = "try-lock"
version = "0.2.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e421abadd41a4225275504ea4d6566923418b7f05506fbc9c0fe86ba7396114b"
[[package]]
name = "tui-input"
version = "0.11.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e5d1733c47f1a217b7deff18730ff7ca4ecafc5771368f715ab072d679a36114"
dependencies = [
"ratatui",
"unicode-width 0.2.0",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "tui-markdown"
version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cf47229087fc49650d095a910a56aaf10c1c64181d042d2c2ba46fc3746ff534"
dependencies = [
"ansi-to-tui",
"itertools 0.14.0",
"pretty_assertions",
"pulldown-cmark",
"ratatui",
"rstest",
"syntect",
"tracing",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "tui-textarea"
version = "0.7.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0a5318dd619ed73c52a9417ad19046724effc1287fb75cdcc4eca1d6ac1acbae"
dependencies = [
"crossterm",
"ratatui",
"unicode-width 0.2.0",
]
[[package]]
name = "unicase"
version = "2.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "75b844d17643ee918803943289730bec8aac480150456169e647ed0b576ba539"
[[package]]
name = "unicode-ident"
version = "1.0.18"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5a5f39404a5da50712a4c1eecf25e90dd62b613502b7e925fd4e4d19b5c96512"
[[package]]
name = "unicode-segmentation"
version = "1.12.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f6ccf251212114b54433ec949fd6a7841275f9ada20dddd2f29e9ceea4501493"
[[package]]
name = "unicode-truncate"
version = "1.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b3644627a5af5fa321c95b9b235a72fd24cd29c648c2c379431e6628655627bf"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"itertools 0.13.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"unicode-segmentation",
"unicode-width 0.1.14",
]
[[package]]
name = "unicode-width"
version = "0.1.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7dd6e30e90baa6f72411720665d41d89b9a3d039dc45b8faea1ddd07f617f6af"
[[package]]
name = "unicode-width"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1fc81956842c57dac11422a97c3b8195a1ff727f06e85c84ed2e8aa277c9a0fd"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "unicode-xid"
version = "0.2.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ebc1c04c71510c7f702b52b7c350734c9ff1295c464a03335b00bb84fc54f853"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "untrusted"
version = "0.9.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8ecb6da28b8a351d773b68d5825ac39017e680750f980f3a1a85cd8dd28a47c1"
[[package]]
name = "url"
version = "2.5.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "32f8b686cadd1473f4bd0117a5d28d36b1ade384ea9b5069a1c40aefed7fda60"
dependencies = [
"form_urlencoded",
"idna",
"percent-encoding",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"serde",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "utf16_iter"
version = "1.0.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c8232dd3cdaed5356e0f716d285e4b40b932ac434100fe9b7e0e8e935b9e6246"
[[package]]
name = "utf8_iter"
version = "1.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b6c140620e7ffbb22c2dee59cafe6084a59b5ffc27a8859a5f0d494b5d52b6be"
[[package]]
name = "utf8parse"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "06abde3611657adf66d383f00b093d7faecc7fa57071cce2578660c9f1010821"
feat: save session transcripts when using Rust CLI (#845) This adds support for saving transcripts when using the Rust CLI. Like the TypeScript CLI, it saves the transcript to `~/.codex/sessions`, though it uses JSONL for the file format (and `.jsonl` for the file extension) so that even if Codex crashes, what was written to the `.jsonl` file should generally still be valid JSONL content.
2025-05-07 13:49:15 -07:00
[[package]]
name = "uuid"
version = "1.16.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "458f7a779bf54acc9f347480ac654f68407d3aab21269a6e3c9f922acd9e2da9"
dependencies = [
"getrandom 0.3.2",
fix: tighten up some logic around session timestamps and ids (#922) * update `SessionConfigured` event to include the UUID for the session * show the UUID in the Rust TUI * use local timestamps in log files instead of UTC * include timestamps in log file names for easier discovery
2025-05-13 19:22:16 -07:00
"serde",
feat: save session transcripts when using Rust CLI (#845) This adds support for saving transcripts when using the Rust CLI. Like the TypeScript CLI, it saves the transcript to `~/.codex/sessions`, though it uses JSONL for the file format (and `.jsonl` for the file extension) so that even if Codex crashes, what was written to the `.jsonl` file should generally still be valid JSONL content.
2025-05-07 13:49:15 -07:00
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "v_frame"
version = "0.3.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d6f32aaa24bacd11e488aa9ba66369c7cd514885742c9fe08cfe85884db3e92b"
dependencies = [
"aligned-vec",
"num-traits",
"wasm-bindgen",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "valuable"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ba73ea9cf16a25df0c8caa16c51acb937d5712a8429db78a3ee29d5dcacd3a65"
[[package]]
name = "vcpkg"
version = "0.2.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "version-compare"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "852e951cb7832cb45cb1169900d19760cfa39b82bc0ea9c0e5a14ae88411c98b"
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "version_check"
version = "0.9.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0b928f33d975fc6ad9f86c8f283853ad26bdd5b10b7f1542aa2fa15e2289105a"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "wait-timeout"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "09ac3b126d3914f9849036f826e054cbabdc8519970b8998ddaf3b5bd3c65f11"
dependencies = [
"libc",
]
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "walkdir"
version = "2.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "29790946404f91d9c5d06f9874efddea1dc06c5efe94541a7d6863108e3a5e4b"
dependencies = [
"same-file",
"winapi-util",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "want"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bfa7760aed19e106de2c7c0b581b509f2f25d3dacaf737cb82ac61bc6d760b0e"
dependencies = [
"try-lock",
]
[[package]]
name = "wasi"
version = "0.11.0+wasi-snapshot-preview1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9c8d87e72b64a3b4db28d11ce29237c246188f4f51057d65a7eab63b7987e423"
[[package]]
name = "wasi"
version = "0.14.2+wasi-0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9683f9a5a998d873c0d21fcbe3c083009670149a8fab228644b8bd36b2c48cb3"
dependencies = [
"wit-bindgen-rt",
]
[[package]]
name = "wasm-bindgen"
version = "0.2.100"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1edc8929d7499fc4e8f0be2262a241556cfc54a0bea223790e71446f2aab1ef5"
dependencies = [
"cfg-if",
"once_cell",
"rustversion",
"wasm-bindgen-macro",
]
[[package]]
name = "wasm-bindgen-backend"
version = "0.2.100"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2f0a0651a5c2bc21487bde11ee802ccaf4c51935d0d3d42a6101f98161700bc6"
dependencies = [
"bumpalo",
"log",
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"wasm-bindgen-shared",
]
[[package]]
name = "wasm-bindgen-futures"
version = "0.4.50"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "555d470ec0bc3bb57890405e5d4322cc9ea83cebb085523ced7be4144dac1e61"
dependencies = [
"cfg-if",
"js-sys",
"once_cell",
"wasm-bindgen",
"web-sys",
]
[[package]]
name = "wasm-bindgen-macro"
version = "0.2.100"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7fe63fc6d09ed3792bd0897b314f53de8e16568c2b3f7982f468c0bf9bd0b407"
dependencies = [
"quote",
"wasm-bindgen-macro-support",
]
[[package]]
name = "wasm-bindgen-macro-support"
version = "0.2.100"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8ae87ea40c9f689fc23f209965b6fb8a99ad69aeeb0231408be24920604395de"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"wasm-bindgen-backend",
"wasm-bindgen-shared",
]
[[package]]
name = "wasm-bindgen-shared"
version = "0.2.100"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1a05d73b933a847d6cccdda8f838a22ff101ad9bf93e33684f39c1f5f0eece3d"
dependencies = [
"unicode-ident",
]
[[package]]
name = "wasm-streams"
version = "0.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "15053d8d85c7eccdbefef60f06769760a563c7f0a9d6902a13d35c7800b0ad65"
dependencies = [
"futures-util",
"js-sys",
"wasm-bindgen",
"wasm-bindgen-futures",
"web-sys",
]
[[package]]
name = "web-sys"
version = "0.3.77"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "33b6dd2ef9186f1f2072e409e99cd22a975331a6b3591b12c764e0e55c60d5d2"
dependencies = [
"js-sys",
"wasm-bindgen",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "weezl"
version = "0.1.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a751b3277700db47d3e574514de2eced5e54dc8a5436a3bf7a0b248b2cee16f3"
feat: introduce support for shell_environment_policy in config.toml (#1061) To date, when handling `shell` and `local_shell` tool calls, we were spawning new processes using the environment inherited from the Codex process itself. This means that the sensitive `OPENAI_API_KEY` that Codex needs to talk to OpenAI models was made available to everything run by `shell` and `local_shell`. While there are cases where that might be useful, it does not seem like a good default. This PR introduces a complex `shell_environment_policy` config option to control the `env` used with these tool calls. It is inevitably a bit complex so that it is possible to override individual components of the policy so without having to restate the entire thing. Details are in the updated `README.md` in this PR, but here is the relevant bit that explains the individual fields of `shell_environment_policy`: | Field | Type | Default | Description | | ------------------------- | -------------------------- | ------- | ----------------------------------------------------------------------------------------------------------------------------------------------- | | `inherit` | string | `core` | Starting template for the environment:<br>`core` (`HOME`, `PATH`, `USER`, …), `all` (clone full parent env), or `none` (start empty). | | `ignore_default_excludes` | boolean | `false` | When `false`, Codex removes any var whose **name** contains `KEY`, `SECRET`, or `TOKEN` (case-insensitive) before other rules run. | | `exclude` | array&lt;string&gt; | `[]` | Case-insensitive glob patterns to drop after the default filter.<br>Examples: `"AWS_*"`, `"AZURE_*"`. | | `set` | table&lt;string,string&gt; | `{}` | Explicit key/value overrides or additions – always win over inherited values. | | `include_only` | array&lt;string&gt; | `[]` | If non-empty, a whitelist of patterns; only variables that match _one_ pattern survive the final step. (Generally used with `inherit = "all"`.) | In particular, note that the default is `inherit = "core"`, so: * if you have extra env variables that you want to inherit from the parent process, use `inherit = "all"` and then specify `include_only` * if you have extra env variables where you want to hardcode the values, the default `inherit = "core"` will work fine, but then you need to specify `set` This configuration is not battle-tested, so we will probably still have to play with it a bit. `core/src/exec_env.rs` has the critical business logic as well as unit tests. Though if nothing else, previous to this change: ``` $ cargo run --bin codex -- debug seatbelt -- printenv OPENAI_API_KEY # ...prints OPENAI_API_KEY... ``` But after this change it does not print anything (as desired). One final thing to call out about this PR is that the `configure_command!` macro we use in `core/src/exec.rs` has to do some complex logic with respect to how it builds up the `env` for the process being spawned under Landlock/seccomp. Specifically, doing `cmd.env_clear()` followed by `cmd.envs(&$env_map)` (which is arguably the most intuitive way to do it) caused the Landlock unit tests to fail because the processes spawned by the unit tests started failing in unexpected ways! If we forgo `env_clear()` in favor of updating env vars one at a time, the tests still pass. The comment in the code talks about this a bit, and while I would like to investigate this more, I need to move on for the moment, but I do plan to come back to it to fully understand what is going on. For example, this suggests that we might not be able to spawn a C program that calls `env_clear()`, which would be...weird. We may still have to fiddle with our Landlock config if that is the case.
2025-05-22 09:51:19 -07:00
[[package]]
name = "wildmatch"
version = "2.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "68ce1ab1f8c62655ebe1350f589c61e505cf94d385bc6a12899442d9081e71fd"
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "winapi"
version = "0.3.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
dependencies = [
"winapi-i686-pc-windows-gnu",
"winapi-x86_64-pc-windows-gnu",
]
[[package]]
name = "winapi-i686-pc-windows-gnu"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "winapi-util"
version = "0.1.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cf221c93e13a30d793f7645a0e7762c55d169dbb0a49671918a2319d289b10bb"
dependencies = [
"windows-sys 0.59.0",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "winapi-x86_64-pc-windows-gnu"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "windows"
version = "0.58.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dd04d41d93c4992d421894c18c8b43496aa748dd4c081bac0dc93eb0489272b6"
dependencies = [
"windows-core 0.58.0",
"windows-targets 0.52.6",
]
[[package]]
name = "windows-core"
version = "0.58.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6ba6d44ec8c2591c134257ce647b7ea6b20335bf6379a27dac5f1641fcf59f99"
dependencies = [
"windows-implement 0.58.0",
"windows-interface 0.58.0",
"windows-result 0.2.0",
"windows-strings 0.1.0",
"windows-targets 0.52.6",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "windows-core"
version = "0.61.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4763c1de310c86d75a878046489e2e5ba02c649d185f21c67d4cf8a56d098980"
dependencies = [
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"windows-implement 0.60.0",
"windows-interface 0.59.1",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"windows-link",
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"windows-result 0.3.2",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"windows-strings 0.4.0",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "windows-implement"
version = "0.58.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2bbd5b46c938e506ecbce286b6628a02171d56153ba733b6c741fc627ec9579b"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "windows-implement"
version = "0.60.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a47fddd13af08290e67f4acabf4b459f647552718f683a7b415d290ac744a836"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "windows-interface"
version = "0.58.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "053c4c462dc91d3b1504c6fe5a726dd15e216ba718e84a0e46a88fbe5ded3515"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "windows-interface"
version = "0.59.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bd9211b69f8dcdfa817bfd14bf1c97c9188afa36f4750130fcdf3f400eca9fa8"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "windows-link"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "76840935b766e1b0a05c0066835fb9ec80071d4c09a16f6bd5f7e655e3c14c38"
[[package]]
name = "windows-registry"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4286ad90ddb45071efd1a66dfa43eb02dd0dfbae1545ad6cc3c51cf34d7e8ba3"
dependencies = [
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
"windows-result 0.3.2",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"windows-strings 0.3.1",
"windows-targets 0.53.0",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "windows-result"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1d1043d8214f791817bab27572aaa8af63732e11bf84aa21a45a78d6c317ae0e"
dependencies = [
"windows-targets 0.52.6",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "windows-result"
version = "0.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c64fd11a4fd95df68efcfee5f44a294fe71b8bc6a91993e2791938abcc712252"
dependencies = [
"windows-link",
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "windows-strings"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4cd9b125c486025df0eabcb585e62173c6c9eddcec5d117d3b6e8c30e2ee4d10"
dependencies = [
"windows-result 0.2.0",
"windows-targets 0.52.6",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "windows-strings"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "87fa48cc5d406560701792be122a10132491cff9d0aeb23583cc2dcafc847319"
dependencies = [
"windows-link",
]
[[package]]
name = "windows-strings"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7a2ba9642430ee452d5a7aa78d72907ebe8cfda358e8cb7918a2050581322f97"
dependencies = [
"windows-link",
]
[[package]]
name = "windows-sys"
version = "0.52.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "282be5f36a8ce781fad8c8ae18fa3f9beff57ec1b52cb3de0789201425d9a33d"
dependencies = [
"windows-targets 0.52.6",
]
[[package]]
name = "windows-sys"
version = "0.59.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e38bc4d79ed67fd075bcc251a1c39b32a1776bbe92e5bef1f0bf1f8c531853b"
dependencies = [
"windows-targets 0.52.6",
]
[[package]]
name = "windows-targets"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9b724f72796e036ab90c1021d4780d4d3d648aca59e491e6b98e725b84e99973"
dependencies = [
"windows_aarch64_gnullvm 0.52.6",
"windows_aarch64_msvc 0.52.6",
"windows_i686_gnu 0.52.6",
"windows_i686_gnullvm 0.52.6",
"windows_i686_msvc 0.52.6",
"windows_x86_64_gnu 0.52.6",
"windows_x86_64_gnullvm 0.52.6",
"windows_x86_64_msvc 0.52.6",
]
[[package]]
name = "windows-targets"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b1e4c7e8ceaaf9cb7d7507c974735728ab453b67ef8f18febdd7c11fe59dca8b"
dependencies = [
"windows_aarch64_gnullvm 0.53.0",
"windows_aarch64_msvc 0.53.0",
"windows_i686_gnu 0.53.0",
"windows_i686_gnullvm 0.53.0",
"windows_i686_msvc 0.53.0",
"windows_x86_64_gnu 0.53.0",
"windows_x86_64_gnullvm 0.53.0",
"windows_x86_64_msvc 0.53.0",
]
[[package]]
name = "windows_aarch64_gnullvm"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "32a4622180e7a0ec044bb555404c800bc9fd9ec262ec147edd5989ccd0c02cd3"
[[package]]
name = "windows_aarch64_gnullvm"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "86b8d5f90ddd19cb4a147a5fa63ca848db3df085e25fee3cc10b39b6eebae764"
[[package]]
name = "windows_aarch64_msvc"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "09ec2a7bb152e2252b53fa7803150007879548bc709c039df7627cabbd05d469"
[[package]]
name = "windows_aarch64_msvc"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c7651a1f62a11b8cbd5e0d42526e55f2c99886c77e007179efff86c2b137e66c"
[[package]]
name = "windows_i686_gnu"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8e9b5ad5ab802e97eb8e295ac6720e509ee4c243f69d781394014ebfe8bbfa0b"
[[package]]
name = "windows_i686_gnu"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c1dc67659d35f387f5f6c479dc4e28f1d4bb90ddd1a5d3da2e5d97b42d6272c3"
[[package]]
name = "windows_i686_gnullvm"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0eee52d38c090b3caa76c563b86c3a4bd71ef1a819287c19d586d7334ae8ed66"
[[package]]
name = "windows_i686_gnullvm"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9ce6ccbdedbf6d6354471319e781c0dfef054c81fbc7cf83f338a4296c0cae11"
[[package]]
name = "windows_i686_msvc"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "240948bc05c5e7c6dabba28bf89d89ffce3e303022809e73deaefe4f6ec56c66"
[[package]]
name = "windows_i686_msvc"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "581fee95406bb13382d2f65cd4a908ca7b1e4c2f1917f143ba16efe98a589b5d"
[[package]]
name = "windows_x86_64_gnu"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "147a5c80aabfbf0c7d901cb5895d1de30ef2907eb21fbbab29ca94c5b08b1a78"
[[package]]
name = "windows_x86_64_gnu"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2e55b5ac9ea33f2fc1716d1742db15574fd6fc8dadc51caab1c16a3d3b4190ba"
[[package]]
name = "windows_x86_64_gnullvm"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "24d5b23dc417412679681396f2b49f3de8c1473deb516bd34410872eff51ed0d"
[[package]]
name = "windows_x86_64_gnullvm"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0a6e035dd0599267ce1ee132e51c27dd29437f63325753051e71dd9e42406c57"
[[package]]
name = "windows_x86_64_msvc"
version = "0.52.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "589f6da84c646204747d1270a2a5661ea66ed1cced2631d546fdfb155959f9ec"
[[package]]
name = "windows_x86_64_msvc"
version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "271414315aff87387382ec3d271b52d7ae78726f5d44ac98b4f4030c91880486"
[[package]]
name = "winnow"
version = "0.7.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "63d3fcd9bba44b03821e7d699eeee959f3126dcc4aa8e4ae18ec617c2a5cea10"
dependencies = [
"memchr",
]
[[package]]
name = "wiremock"
version = "0.6.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "101681b74cd87b5899e87bcf5a64e83334dd313fcd3053ea72e6dba18928e301"
dependencies = [
"assert-json-diff",
"async-trait",
"base64 0.22.1",
"deadpool",
"futures",
"http",
"http-body-util",
"hyper",
"hyper-util",
"log",
"once_cell",
"regex",
"serde",
"serde_json",
"tokio",
"url",
]
[[package]]
name = "wit-bindgen-rt"
version = "0.39.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6f42320e61fe2cfd34354ecb597f86f413484a798ba44a8ca1165c58d42da6c1"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"bitflags 2.9.0",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "write16"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d1890f4022759daae28ed4fe62859b1236caebfc61ede2f63ed4e695f3f6d936"
[[package]]
name = "writeable"
version = "0.5.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1e9df38ee2d2c3c5948ea468a8406ff0db0b29ae1ffde1bcf20ef305bcc95c51"
feat: introduce the use of tui-markdown (#851) This introduces the use of the `tui-markdown` crate to parse an assistant message as Markdown and style it using ANSI for a better user experience. As shown in the screenshot below, it has support for syntax highlighting for _tagged_ fenced code blocks: <img width="907" alt="image" src="https://github.com/user-attachments/assets/900dc229-80bb-46e8-b1bb-efee4c70ba3c" /> That said, `tui-markdown` is not as configurable (or stylish!) as https://www.npmjs.com/package/marked-terminal, which is what we use in the TypeScript CLI. In particular: * The styles are hardcoded and `tui_markdown::from_str()` does not take any options whatsoever. It uses "bold white" for inline code style which does not stand out as much as the yellow used by `marked-terminal`: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L464 I asked Codex to take a first pass at this and it came up with: https://github.com/joshka/tui-markdown/pull/80 * If a fenced code block is not tagged, then it does not get highlighted. I would rather add some logic here: https://github.com/joshka/tui-markdown/blob/65402cbda70325f34e7ddf6fe1ec629bcd9459cf/tui-markdown/src/lib.rs#L262 that uses something like https://pypi.org/project/guesslang/ to examine the value of `text` and try to use the appropriate syntax highlighter. * When we have a fenced code block, we do not want to show the opening and closing triple backticks in the output. To unblock ourselves, we might want to bundle our own fork of `tui-markdown` temporarily until we figure out what the shape of the API should be and then try to upstream it.
2025-05-07 10:46:32 -07:00
[[package]]
name = "yaml-rust"
version = "0.4.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56c1936c4cc7a1c9ab21a1ebb602eb942ba868cbd44a99cb7cdc5892335e1c85"
dependencies = [
"linked-hash-map",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "yansi"
version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cfe53a6657fd280eaa890a3bc59152892ffa3e30101319d168b781ed6529b049"
[[package]]
name = "yoke"
version = "0.7.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "120e6aef9aa629e3d4f52dc8cc43a015c7724194c97dfaf45180d2daf2b77f40"
dependencies = [
"serde",
"stable_deref_trait",
"yoke-derive",
"zerofrom",
]
[[package]]
name = "yoke-derive"
version = "0.7.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2380878cad4ac9aac1e2435f3eb4020e8374b5f13c296cb75b4620ff8e229154"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"synstructure",
]
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
[[package]]
name = "zerocopy"
version = "0.7.35"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1b9b4fd18abc82b8136838da5d50bae7bdea537c574d8dc1a34ed098d6c166f0"
dependencies = [
"zerocopy-derive 0.7.35",
]
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
[[package]]
name = "zerocopy"
version = "0.8.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2586fea28e186957ef732a5f8b3be2da217d65c5969d4b1e17f973ebbe876879"
dependencies = [
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"zerocopy-derive 0.8.24",
]
[[package]]
name = "zerocopy-derive"
version = "0.7.35"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fa4f8080344d4671fb4e831a13ad1e68092748387dfc4f55e356242fae12ce3e"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "zerocopy-derive"
version = "0.8.24"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a996a8f63c5c4448cd959ac1bab0aaa3306ccfd060472f85943ee0750f0169be"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
[[package]]
name = "zerofrom"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "50cc42e0333e05660c3587f3bf9d0478688e15d870fab3346451ce7f8c9fbea5"
dependencies = [
"zerofrom-derive",
]
[[package]]
name = "zerofrom-derive"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d71e5d6e06ab090c67b5e44993ec16b72dcbaabc526db883a360057678b48502"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
"synstructure",
]
[[package]]
name = "zeroize"
version = "1.8.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ced3678a2879b30306d323f4542626697a464a97c0a07c9aebf7ebca65cd4dde"
[[package]]
name = "zerovec"
version = "0.10.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "aa2b893d79df23bfb12d5461018d408ea19dfafe76c2c7ef6d4eba614f8ff079"
dependencies = [
"yoke",
"zerofrom",
"zerovec-derive",
]
[[package]]
name = "zerovec-derive"
version = "0.10.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6eafa6dfb17584ea3e2bd6e76e0cc15ad7af12b09abdd1ca55961bed9b1063c6"
dependencies = [
"proc-macro2",
"quote",
feat: introduce codex_execpolicy crate for defining "safe" commands (#634) As described in detail in `codex-rs/execpolicy/README.md` introduced in this PR, `execpolicy` is a tool that lets you define a set of _patterns_ used to match [`execv(3)`](https://linux.die.net/man/3/execv) invocations. When a pattern is matched, `execpolicy` returns the parsed version in a structured form that is amenable to static analysis. The primary use case is to define patterns match commands that should be auto-approved by a tool such as Codex. This supports a richer pattern matching mechanism that the sort of prefix-matching we have done to date, e.g.: https://github.com/openai/codex/blob/5e40d9d2211737f46136610497bcd9a8271009e0/codex-cli/src/approvals.ts#L333-L354 Note we are still playing with the API and the `system_path` option in particular still needs some work.
2025-04-24 17:14:47 -07:00
"syn 2.0.100",
feat: initial import of Rust implementation of Codex CLI in codex-rs/ (#629) As stated in `codex-rs/README.md`: Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible. To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits: - The CLI compiles to small, standalone, platform-specific binaries. - Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux. - No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance. Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implmentation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
2025-04-24 13:31:40 -07:00
]
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
[[package]]
name = "zune-core"
version = "0.4.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3f423a2c17029964870cfaabb1f13dfab7d092a62a29a89264f4d36990ca414a"
[[package]]
name = "zune-inflate"
version = "0.2.54"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "73ab332fe2f6680068f3582b16a24f90ad7096d5d39b974d1c0aff0125116f02"
dependencies = [
"simd-adler32",
]
[[package]]
name = "zune-jpeg"
version = "0.4.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "99a5bab8d7dedf81405c4bb1f2b83ea057643d9cb28778cea9eecddeedd2e028"
dependencies = [
"zune-core",
]
Reference in New Issue Copy Permalink