Files
llmx/codex-rs/mcp-server/src/codex_tool_runner.rs

311 lines
12 KiB
Rust
Raw Normal View History

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
//! Asynchronous worker that executes a **Codex** tool-call inside a spawned
//! Tokio task. Separated from `message_processor.rs` to keep that file small
//! and to make future feature-growth easier to manage.
use std::path::PathBuf;
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
use std::sync::Arc;
use codex_core::Codex;
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
use codex_core::codex_wrapper::init_codex;
use codex_core::config::Config as CodexConfig;
use codex_core::protocol::AgentMessageEvent;
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
use codex_core::protocol::EventMsg;
use codex_core::protocol::ExecApprovalRequestEvent;
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
use codex_core::protocol::InputItem;
use codex_core::protocol::Op;
use codex_core::protocol::ReviewDecision;
use codex_core::protocol::Submission;
use codex_core::protocol::TaskCompleteEvent;
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
use mcp_types::CallToolResult;
use mcp_types::ContentBlock;
use mcp_types::ElicitRequest;
use mcp_types::ElicitRequestParamsRequestedSchema;
use mcp_types::JSONRPCErrorError;
use mcp_types::ModelContextProtocolRequest;
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
use mcp_types::RequestId;
use mcp_types::TextContent;
use serde::Deserialize;
use serde::Serialize;
use serde_json::json;
use tracing::error;
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
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
use crate::outgoing_message::OutgoingMessageSender;
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
const INVALID_PARAMS_ERROR_CODE: i64 = -32602;
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
/// Run a complete Codex session and stream events back to the client.
///
/// On completion (success or error) the function sends the appropriate
/// `tools/call` response so the LLM can continue the conversation.
pub async fn run_codex_tool_session(
id: RequestId,
initial_prompt: String,
config: CodexConfig,
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing: Arc<OutgoingMessageSender>,
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
) {
let (codex, first_event, _ctrl_c) = match init_codex(config).await {
Ok(res) => res,
Err(e) => {
let result = CallToolResult {
content: vec![ContentBlock::TextContent(TextContent {
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
r#type: "text".to_string(),
text: format!("Failed to start Codex session: {e}"),
annotations: None,
})],
is_error: Some(true),
structured_content: None,
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
};
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing.send_response(id.clone(), result.into()).await;
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
return;
}
};
let codex = Arc::new(codex);
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
// Send initial SessionConfigured event.
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing.send_event_as_notification(&first_event).await;
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
// Use the original MCP request ID as the `sub_id` for the Codex submission so that
// any events emitted for this tool-call can be correlated with the
// originating `tools/call` request.
let sub_id = match &id {
RequestId::String(s) => s.clone(),
RequestId::Integer(n) => n.to_string(),
};
let submission = Submission {
id: sub_id.clone(),
op: Op::UserInput {
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
items: vec![InputItem::Text {
text: initial_prompt.clone(),
}],
},
};
if let Err(e) = codex.submit_with_id(submission).await {
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
tracing::error!("Failed to submit initial prompt: {e}");
}
// Stream events until the task needs to pause for user interaction or
// completes.
loop {
match codex.next_event().await {
Ok(event) => {
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing.send_event_as_notification(&event).await;
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
match event.msg {
EventMsg::ExecApprovalRequest(ExecApprovalRequestEvent {
command,
cwd,
reason: _,
}) => {
let escaped_command = shlex::try_join(command.iter().map(|s| s.as_str()))
.unwrap_or_else(|_| command.join(" "));
test: add integration test for MCP server (#1633) This PR introduces a single integration test for `cargo mcp`, though it also introduces a number of reusable components so that it should be easier to introduce more integration tests going forward. The new test is introduced in `codex-rs/mcp-server/tests/elicitation.rs` and the reusable pieces are in `codex-rs/mcp-server/tests/common`. The test itself verifies new functionality around elicitations introduced in https://github.com/openai/codex/pull/1623 (and the fix introduced in https://github.com/openai/codex/pull/1629) by doing the following: - starts a mock model provider with canned responses for `/v1/chat/completions` - starts the MCP server with a `config.toml` to use that model provider (and `approval_policy = "untrusted"`) - sends the `codex` tool call which causes the mock model provider to request a shell call for `git init` - the MCP server sends an elicitation to the client to approve the request - the client replies to the elicitation with `"approved"` - the MCP server runs the command and re-samples the model, getting a `"finish_reason": "stop"` - in turn, the MCP server sends the final response to the original `codex` tool call - verifies that `git init` ran as expected To test: ``` cargo test shell_command_approval_triggers_elicitation ``` In writing this test, I discovered that `ExecApprovalResponse` does not conform to `ElicitResult`, so I added a TODO to fix that, since I think that should be updated in a separate PR. As it stands, this PR does not update any business logic, though it does make a number of members of the `mcp-server` crate `pub` so they can be used in the test. One additional learning from this PR is that `std::process::Command::cargo_bin()` from the `assert_cmd` trait is only available for `std::process::Command`, but we really want to use `tokio::process::Command` so that everything is async and we can leverage utilities like `tokio::time::timeout()`. The trick I came up with was to use `cargo_bin()` to locate the program, and then to use `std::process::Command::get_program()` when constructing the `tokio::process::Command`.
2025-07-21 10:27:07 -07:00
let message = format!(
"Allow Codex to run `{escaped_command}` in `{cwd}`?",
cwd = cwd.to_string_lossy()
);
let params = ExecApprovalElicitRequestParams {
message,
requested_schema: ElicitRequestParamsRequestedSchema {
r#type: "object".to_string(),
properties: json!({}),
required: None,
},
codex_elicitation: "exec-approval".to_string(),
codex_mcp_tool_call_id: sub_id.clone(),
codex_event_id: event.id.clone(),
codex_command: command,
codex_cwd: cwd,
};
let params_json = match serde_json::to_value(&params) {
Ok(value) => value,
Err(err) => {
let message = format!(
"Failed to serialize ExecApprovalElicitRequestParams: {err}"
);
tracing::error!("{message}");
outgoing
.send_error(
id.clone(),
JSONRPCErrorError {
code: INVALID_PARAMS_ERROR_CODE,
message,
data: None,
},
)
.await;
continue;
}
};
let on_response = outgoing
.send_request(ElicitRequest::METHOD, Some(params_json))
.await;
// Listen for the response on a separate task so we do
// not block the main loop of this function.
{
let codex = codex.clone();
let event_id = event.id.clone();
tokio::spawn(async move {
on_exec_approval_response(event_id, on_response, codex).await;
});
}
// Continue, don't break so the session continues.
continue;
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
}
EventMsg::ApplyPatchApprovalRequest(_) => {
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
let result = CallToolResult {
content: vec![ContentBlock::TextContent(TextContent {
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
r#type: "text".to_string(),
text: "PATCH_APPROVAL_REQUIRED".to_string(),
annotations: None,
})],
is_error: None,
structured_content: None,
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
};
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing.send_response(id.clone(), result.into()).await;
// Continue, don't break so the session continues.
continue;
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
}
EventMsg::TaskComplete(TaskCompleteEvent { last_agent_message }) => {
let text = match last_agent_message {
Some(msg) => msg.clone(),
None => "".to_string(),
};
let result = CallToolResult {
content: vec![ContentBlock::TextContent(TextContent {
r#type: "text".to_string(),
text,
annotations: None,
})],
is_error: None,
structured_content: None,
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
};
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing.send_response(id.clone(), result.into()).await;
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
break;
}
EventMsg::SessionConfigured(_) => {
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
tracing::error!("unexpected SessionConfigured event");
}
EventMsg::AgentMessageDelta(_) => {
// TODO: think how we want to support this in the MCP
}
EventMsg::AgentReasoningDelta(_) => {
// TODO: think how we want to support this in the MCP
}
EventMsg::AgentMessage(AgentMessageEvent { .. }) => {
// TODO: think how we want to support this in the MCP
}
EventMsg::Error(_)
| EventMsg::TaskStarted
feat: show number of tokens remaining in UI (#1388) When using the OpenAI Responses API, we now record the `usage` field for a `"response.completed"` event, which includes metrics about the number of tokens consumed. We also introduce `openai_model_info.rs`, which includes current data about the most common OpenAI models available via the API (specifically `context_window` and `max_output_tokens`). If Codex does not recognize the model, you can set `model_context_window` and `model_max_output_tokens` explicitly in `config.toml`. When then introduce a new event type to `protocol.rs`, `TokenCount`, which includes the `TokenUsage` for the most recent turn. Finally, we update the TUI to record the running sum of tokens used so the percentage of available context window remaining can be reported via the placeholder text for the composer: ![Screenshot 2025-06-25 at 11 20 55 PM](https://github.com/user-attachments/assets/6fd6982f-7247-4f14-84b2-2e600cb1fd49) We could certainly get much fancier with this (such as reporting the estimated cost of the conversation), but for now, we are just trying to achieve feature parity with the TypeScript CLI. Though arguably this improves upon the TypeScript CLI, as the TypeScript CLI uses heuristics to estimate the number of tokens used rather than using the `usage` information directly: https://github.com/openai/codex/blob/296996d74e345b1b05d8c3451a06ace21c5ada96/codex-cli/src/utils/approximate-tokens-used.ts#L3-L16 Fixes https://github.com/openai/codex/issues/1242
2025-06-25 23:31:11 -07:00
| EventMsg::TokenCount(_)
| EventMsg::AgentReasoning(_)
| EventMsg::McpToolCallBegin(_)
| EventMsg::McpToolCallEnd(_)
| EventMsg::ExecCommandBegin(_)
| EventMsg::ExecCommandEnd(_)
| EventMsg::BackgroundEvent(_)
| EventMsg::PatchApplyBegin(_)
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
| EventMsg::PatchApplyEnd(_)
| EventMsg::GetHistoryEntryResponse(_) => {
// For now, we do not do anything extra for these
// events. Note that
// send(codex_event_to_notification(&event)) above has
// already dispatched these events as notifications,
// though we may want to do give different treatment to
// individual events in the future.
}
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
}
}
Err(e) => {
let result = CallToolResult {
content: vec![ContentBlock::TextContent(TextContent {
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
r#type: "text".to_string(),
text: format!("Codex runtime error: {e}"),
annotations: None,
})],
is_error: Some(true),
// TODO(mbolin): Could present the error in a more
// structured way.
structured_content: None,
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
};
chore: introduce OutgoingMessageSender (#1622) Previous to this change, `MessageProcessor` had a `tokio::sync::mpsc::Sender<JSONRPCMessage>` as an abstraction for server code to send a message down to the MCP client. Because `Sender` is cheap to `clone()`, it was straightforward to make it available to tasks scheduled with `tokio::task::spawn()`. This worked well when we were only sending notifications or responses back down to the client, but we want to add support for sending elicitations in #1623, which means that we need to be able to send _requests_ to the client, and now we need a bit of centralization to ensure all request ids are unique. To that end, this PR introduces `OutgoingMessageSender`, which houses the existing `Sender<OutgoingMessage>` as well as an `AtomicI64` to mint out new, unique request ids. It has methods like `send_request()` and `send_response()` so that callers do not have to deal with `JSONRPCMessage` directly, as having to set the `jsonrpc` for each message was a bit tedious (this cleans up `codex_tool_runner.rs` quite a bit). We do not have `OutgoingMessageSender` implement `Clone` because it is important that the `AtomicI64` is shared across all users of `OutgoingMessageSender`. As such, `Arc<OutgoingMessageSender>` must be used instead, as it is frequently shared with new tokio tasks. As part of this change, we update `message_processor.rs` to embrace `await`, though we must be careful that no individual handler blocks the main loop and prevents other messages from being handled. --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1622). * #1623 * __->__ #1622 * #1621 * #1620
2025-07-19 00:30:56 -04:00
outgoing.send_response(id.clone(), result.into()).await;
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
break;
}
}
}
}
async fn on_exec_approval_response(
event_id: String,
receiver: tokio::sync::oneshot::Receiver<mcp_types::Result>,
codex: Arc<Codex>,
) {
let response = receiver.await;
let value = match response {
Ok(value) => value,
Err(err) => {
error!("request failed: {err:?}");
return;
}
};
// Try to deserialize `value` and then make the appropriate call to `codex`.
let response = match serde_json::from_value::<ExecApprovalResponse>(value) {
Ok(response) => response,
Err(err) => {
error!("failed to deserialize ExecApprovalResponse: {err}");
// If we cannot deserialize the response, we deny the request to be
// conservative.
ExecApprovalResponse {
decision: ReviewDecision::Denied,
}
}
};
if let Err(err) = codex
.submit(Op::ExecApproval {
id: event_id,
decision: response.decision,
})
.await
{
error!("failed to submit ExecApproval: {err}");
}
}
test: add integration test for MCP server (#1633) This PR introduces a single integration test for `cargo mcp`, though it also introduces a number of reusable components so that it should be easier to introduce more integration tests going forward. The new test is introduced in `codex-rs/mcp-server/tests/elicitation.rs` and the reusable pieces are in `codex-rs/mcp-server/tests/common`. The test itself verifies new functionality around elicitations introduced in https://github.com/openai/codex/pull/1623 (and the fix introduced in https://github.com/openai/codex/pull/1629) by doing the following: - starts a mock model provider with canned responses for `/v1/chat/completions` - starts the MCP server with a `config.toml` to use that model provider (and `approval_policy = "untrusted"`) - sends the `codex` tool call which causes the mock model provider to request a shell call for `git init` - the MCP server sends an elicitation to the client to approve the request - the client replies to the elicitation with `"approved"` - the MCP server runs the command and re-samples the model, getting a `"finish_reason": "stop"` - in turn, the MCP server sends the final response to the original `codex` tool call - verifies that `git init` ran as expected To test: ``` cargo test shell_command_approval_triggers_elicitation ``` In writing this test, I discovered that `ExecApprovalResponse` does not conform to `ElicitResult`, so I added a TODO to fix that, since I think that should be updated in a separate PR. As it stands, this PR does not update any business logic, though it does make a number of members of the `mcp-server` crate `pub` so they can be used in the test. One additional learning from this PR is that `std::process::Command::cargo_bin()` from the `assert_cmd` trait is only available for `std::process::Command`, but we really want to use `tokio::process::Command` so that everything is async and we can leverage utilities like `tokio::time::timeout()`. The trick I came up with was to use `cargo_bin()` to locate the program, and then to use `std::process::Command::get_program()` when constructing the `tokio::process::Command`.
2025-07-21 10:27:07 -07:00
// TODO(mbolin): ExecApprovalResponse does not conform to ElicitResult. See:
// - https://github.com/modelcontextprotocol/modelcontextprotocol/blob/f962dc1780fa5eed7fb7c8a0232f1fc83ef220cd/schema/2025-06-18/schema.json#L617-L636
// - https://modelcontextprotocol.io/specification/draft/client/elicitation#protocol-messages
// It should have "action" and "content" fields.
#[derive(Debug, Serialize, Deserialize)]
pub struct ExecApprovalResponse {
pub decision: ReviewDecision,
}
/// Conforms to [`mcp_types::ElicitRequestParams`] so that it can be used as the
/// `params` field of an [`mcp_types::ElicitRequest`].
#[derive(Debug, Serialize)]
test: add integration test for MCP server (#1633) This PR introduces a single integration test for `cargo mcp`, though it also introduces a number of reusable components so that it should be easier to introduce more integration tests going forward. The new test is introduced in `codex-rs/mcp-server/tests/elicitation.rs` and the reusable pieces are in `codex-rs/mcp-server/tests/common`. The test itself verifies new functionality around elicitations introduced in https://github.com/openai/codex/pull/1623 (and the fix introduced in https://github.com/openai/codex/pull/1629) by doing the following: - starts a mock model provider with canned responses for `/v1/chat/completions` - starts the MCP server with a `config.toml` to use that model provider (and `approval_policy = "untrusted"`) - sends the `codex` tool call which causes the mock model provider to request a shell call for `git init` - the MCP server sends an elicitation to the client to approve the request - the client replies to the elicitation with `"approved"` - the MCP server runs the command and re-samples the model, getting a `"finish_reason": "stop"` - in turn, the MCP server sends the final response to the original `codex` tool call - verifies that `git init` ran as expected To test: ``` cargo test shell_command_approval_triggers_elicitation ``` In writing this test, I discovered that `ExecApprovalResponse` does not conform to `ElicitResult`, so I added a TODO to fix that, since I think that should be updated in a separate PR. As it stands, this PR does not update any business logic, though it does make a number of members of the `mcp-server` crate `pub` so they can be used in the test. One additional learning from this PR is that `std::process::Command::cargo_bin()` from the `assert_cmd` trait is only available for `std::process::Command`, but we really want to use `tokio::process::Command` so that everything is async and we can leverage utilities like `tokio::time::timeout()`. The trick I came up with was to use `cargo_bin()` to locate the program, and then to use `std::process::Command::get_program()` when constructing the `tokio::process::Command`.
2025-07-21 10:27:07 -07:00
pub struct ExecApprovalElicitRequestParams {
// These fields are required so that `params`
// conforms to ElicitRequestParams.
test: add integration test for MCP server (#1633) This PR introduces a single integration test for `cargo mcp`, though it also introduces a number of reusable components so that it should be easier to introduce more integration tests going forward. The new test is introduced in `codex-rs/mcp-server/tests/elicitation.rs` and the reusable pieces are in `codex-rs/mcp-server/tests/common`. The test itself verifies new functionality around elicitations introduced in https://github.com/openai/codex/pull/1623 (and the fix introduced in https://github.com/openai/codex/pull/1629) by doing the following: - starts a mock model provider with canned responses for `/v1/chat/completions` - starts the MCP server with a `config.toml` to use that model provider (and `approval_policy = "untrusted"`) - sends the `codex` tool call which causes the mock model provider to request a shell call for `git init` - the MCP server sends an elicitation to the client to approve the request - the client replies to the elicitation with `"approved"` - the MCP server runs the command and re-samples the model, getting a `"finish_reason": "stop"` - in turn, the MCP server sends the final response to the original `codex` tool call - verifies that `git init` ran as expected To test: ``` cargo test shell_command_approval_triggers_elicitation ``` In writing this test, I discovered that `ExecApprovalResponse` does not conform to `ElicitResult`, so I added a TODO to fix that, since I think that should be updated in a separate PR. As it stands, this PR does not update any business logic, though it does make a number of members of the `mcp-server` crate `pub` so they can be used in the test. One additional learning from this PR is that `std::process::Command::cargo_bin()` from the `assert_cmd` trait is only available for `std::process::Command`, but we really want to use `tokio::process::Command` so that everything is async and we can leverage utilities like `tokio::time::timeout()`. The trick I came up with was to use `cargo_bin()` to locate the program, and then to use `std::process::Command::get_program()` when constructing the `tokio::process::Command`.
2025-07-21 10:27:07 -07:00
pub message: String,
#[serde(rename = "requestedSchema")]
test: add integration test for MCP server (#1633) This PR introduces a single integration test for `cargo mcp`, though it also introduces a number of reusable components so that it should be easier to introduce more integration tests going forward. The new test is introduced in `codex-rs/mcp-server/tests/elicitation.rs` and the reusable pieces are in `codex-rs/mcp-server/tests/common`. The test itself verifies new functionality around elicitations introduced in https://github.com/openai/codex/pull/1623 (and the fix introduced in https://github.com/openai/codex/pull/1629) by doing the following: - starts a mock model provider with canned responses for `/v1/chat/completions` - starts the MCP server with a `config.toml` to use that model provider (and `approval_policy = "untrusted"`) - sends the `codex` tool call which causes the mock model provider to request a shell call for `git init` - the MCP server sends an elicitation to the client to approve the request - the client replies to the elicitation with `"approved"` - the MCP server runs the command and re-samples the model, getting a `"finish_reason": "stop"` - in turn, the MCP server sends the final response to the original `codex` tool call - verifies that `git init` ran as expected To test: ``` cargo test shell_command_approval_triggers_elicitation ``` In writing this test, I discovered that `ExecApprovalResponse` does not conform to `ElicitResult`, so I added a TODO to fix that, since I think that should be updated in a separate PR. As it stands, this PR does not update any business logic, though it does make a number of members of the `mcp-server` crate `pub` so they can be used in the test. One additional learning from this PR is that `std::process::Command::cargo_bin()` from the `assert_cmd` trait is only available for `std::process::Command`, but we really want to use `tokio::process::Command` so that everything is async and we can leverage utilities like `tokio::time::timeout()`. The trick I came up with was to use `cargo_bin()` to locate the program, and then to use `std::process::Command::get_program()` when constructing the `tokio::process::Command`.
2025-07-21 10:27:07 -07:00
pub requested_schema: ElicitRequestParamsRequestedSchema,
// These are additional fields the client can use to
// correlate the request with the codex tool call.
test: add integration test for MCP server (#1633) This PR introduces a single integration test for `cargo mcp`, though it also introduces a number of reusable components so that it should be easier to introduce more integration tests going forward. The new test is introduced in `codex-rs/mcp-server/tests/elicitation.rs` and the reusable pieces are in `codex-rs/mcp-server/tests/common`. The test itself verifies new functionality around elicitations introduced in https://github.com/openai/codex/pull/1623 (and the fix introduced in https://github.com/openai/codex/pull/1629) by doing the following: - starts a mock model provider with canned responses for `/v1/chat/completions` - starts the MCP server with a `config.toml` to use that model provider (and `approval_policy = "untrusted"`) - sends the `codex` tool call which causes the mock model provider to request a shell call for `git init` - the MCP server sends an elicitation to the client to approve the request - the client replies to the elicitation with `"approved"` - the MCP server runs the command and re-samples the model, getting a `"finish_reason": "stop"` - in turn, the MCP server sends the final response to the original `codex` tool call - verifies that `git init` ran as expected To test: ``` cargo test shell_command_approval_triggers_elicitation ``` In writing this test, I discovered that `ExecApprovalResponse` does not conform to `ElicitResult`, so I added a TODO to fix that, since I think that should be updated in a separate PR. As it stands, this PR does not update any business logic, though it does make a number of members of the `mcp-server` crate `pub` so they can be used in the test. One additional learning from this PR is that `std::process::Command::cargo_bin()` from the `assert_cmd` trait is only available for `std::process::Command`, but we really want to use `tokio::process::Command` so that everything is async and we can leverage utilities like `tokio::time::timeout()`. The trick I came up with was to use `cargo_bin()` to locate the program, and then to use `std::process::Command::get_program()` when constructing the `tokio::process::Command`.
2025-07-21 10:27:07 -07:00
pub codex_elicitation: String,
pub codex_mcp_tool_call_id: String,
pub codex_event_id: String,
pub codex_command: Vec<String>,
pub codex_cwd: PathBuf,
}