As discovered in https://github.com/openai/codex/issues/1365, the Azure
provider needs to be able to specify `api-version` as a query param, so
this PR introduces a generic `query_params` option to the
`model_providers` config so that an Azure provider can be defined as
follows:
```toml
[model_providers.azure]
name = "Azure"
base_url = "https://YOUR_PROJECT_NAME.openai.azure.com/openai"
env_key = "AZURE_OPENAI_API_KEY"
query_params = { api-version = "2025-04-01-preview" }
```
This PR also updates the docs with this example.
While here, we also update `wire_api` to default to `"chat"`, as that is
likely the common case for someone defining an external provider.
Fixes https://github.com/openai/codex/issues/1365.
Looking at existing releases such as
https://github.com/openai/codex/releases/tag/codex-rs-b289c9207090b2e27494545d7b5404e063bd86f3-1-rust-v0.1.0-alpha.4,
the `.tar.gz` for the source code still seems to have `0.0.0` as the
`version` in `codex-rs/Cargo.toml` instead of what the tag seems to say
it should have:
b289c92070/codex-rs/Cargo.toml (L21)
ChatGPT claims:
> When GitHub generates the Source code (tar.gz) archive for a tag:
• It uses the commit the tag points to.
• But in some cases (e.g., shallow clones, GitHub CI, or local tools
that only clone the default branch), that commit may not be included,
and you might get an outdated view or nothing at all depending on how
it’s fetched.
Trying this recommended fix.
This is a small quality-of-life feature, the addition of
`--compute-indices` to the CLI, which, if enabled, will compute and set
the `indices` field for each `FileMatch` returned by `run()`. Note we
only bother to compute `indices` once we have the top N results because
there could be a lot of intermediate "top N" results during the search
that are ultimately discarded.
When set, the indices are included in the JSON output when `--json` is
specified and the matching indices are displayed in bold when `--json`
is not specified.
Introduces support for `@` to trigger a fuzzy-filename search in the
composer. Under the hood, this leverages
https://crates.io/crates/nucleo-matcher to do the fuzzy matching and
https://crates.io/crates/ignore to build up the list of file candidates
(so that it respects `.gitignore`).
For simplicity (at least for now), we do not do any caching between
searches like VS Code does for its file search:
1d89ed699b/src/vs/workbench/services/search/node/rawSearchService.ts (L212-L218)
Because we do not do any caching, I saw queries take up to three seconds
on large repositories with hundreds of thousands of files. To that end,
we do not perform searches synchronously on each keystroke, but instead
dispatch an event to do the search on a background thread that
asynchronously reports back to the UI when the results are available.
This is largely handled by the `FileSearchManager` introduced in this
PR, which also has logic for debouncing requests so there is at most one
search in flight at a time.
While we could potentially polish and tune this feature further, it may
already be overengineered for how it will be used, in practice, so we
can improve things going forward if it turns out that this is not "good
enough" in the wild.
Note this feature does not work like `@` in the TypeScript CLI, which
was more like directory-based tab completion. In the Rust CLI, `@`
triggers a full-repo fuzzy-filename search.
Fixes https://github.com/openai/codex/issues/1261.
Update `run()` to take `cancel_flag: Arc<AtomicBool>` that the worker
threads will periodically check to see if it is `true`, exiting early
(and returning empty results) if so.
As we are [close to releasing the Rust CLI
beta](https://github.com/openai/codex/discussions/1405), for the moment,
let's take a more neutral stance on what it takes to be a "built-in"
provider.
* For example, there seems to be a discrepancy around what the "right"
configuration for Gemini is: https://github.com/openai/codex/pull/881
* And while the current list of "built-in" providers are all arguably
"well-known" names, this raises a question of what to do about
potentially less familiar providers, such as
https://github.com/openai/codex/pull/1142. Do we just accept every pull
request like this, or is there some criteria a provider has to meet to
"qualify" to be bundled with Codex CLI?
I think that if we can establish clear ground rules for being a built-in
provider, then we can bring this back. But until then, I would rather
take a minimalist approach because if we decided to reverse our position
later, it would break folks who were depending on the presence of the
built-in providers.
Adds support for a `/diff` command comparable to the one available in
the TypeScript CLI.
<img width="1103" alt="Screenshot 2025-06-26 at 12 31 33 PM"
src="https://github.com/user-attachments/assets/5dc646ca-301f-41ff-92a7-595c68db64b6"
/>
While here, changed the `SlashCommand` enum so the declared variant
order is the order the commands appear in the popup menu. This way,
`/toggle-mouse-mode` is listed last, as it is the least likely to be
used.
Fixes https://github.com/openai/codex/issues/1253.
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:
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:
296996d74e/codex-cli/src/utils/approximate-tokens-used.ts (L3-L16)
Fixes https://github.com/openai/codex/issues/1242
This PR reworks `assess_command_safety()` so that the combination of
`AskForApproval::Never` and `SandboxPolicy::DangerFullAccess` ensures
that commands are run without _any_ sandbox and the user should never be
prompted. In turn, it adds support for a new
`--dangerously-bypass-approvals-and-sandbox` flag (that cannot be used
with `--approval-policy` or `--full-auto`) that sets both of those
options.
Fixes https://github.com/openai/codex/issues/1254
For the `approval_policy` config option, renames `unless-allow-listed`
to `untrusted`. In general, when it comes to exec'ing commands, I think
"trusted" is a more accurate term than "safe."
Also drops the `AskForApproval::AutoEdit` variant, as we were not really
making use of it, anyway.
Fixes https://github.com/openai/codex/issues/1250.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1378).
* #1379
* __->__ #1378
Apparently `just` was added to `apt` in Ubuntu 24, so this required
updating the Ubuntu version in the `Dockerfile` to make it so we could
simply `apt install just`.
Though then that caused a conflict with the custom `dev` user we were
using, though the end result seems simpler since now we just use the
default `ubuntu` user provided by Ubuntu 24.
This is a major redesign of how sandbox configuration works and aims to
fix https://github.com/openai/codex/issues/1248. Specifically, it
replaces `sandbox_permissions` in `config.toml` (and the
`-s`/`--sandbox-permission` CLI flags) with a "table" with effectively
three variants:
```toml
# Safest option: full disk is read-only, but writes and network access are disallowed.
[sandbox]
mode = "read-only"
# The cwd of the Codex task is writable, as well as $TMPDIR on macOS.
# writable_roots can be used to specify additional writable folders.
[sandbox]
mode = "workspace-write"
writable_roots = [] # Optional, defaults to the empty list.
network_access = false # Optional, defaults to false.
# Disable sandboxing: use at your own risk!!!
[sandbox]
mode = "danger-full-access"
```
This should make sandboxing easier to reason about. While we have
dropped support for `-s`, the way it works now is:
- no flags => `read-only`
- `--full-auto` => `workspace-write`
- currently, there is no way to specify `danger-full-access` via a CLI
flag, but we will revisit that as part of
https://github.com/openai/codex/issues/1254
Outstanding issue:
- As noted in the `TODO` on `SandboxPolicy::is_unrestricted()`, we are
still conflating sandbox preferences with approval preferences in that
case, which needs to be cleaned up.
- Use Responses API for Azure provider endpoints
- Added a unit test to catch regression on the change from
`/chat/completions` to `/responses`
- Updated the default AOAI api version from `2025-03-01-preview` to
`2025-04-01-preview` to avoid user/400 errors due to missing summary
support in the March API version.
- Changes have been tested locally on AOAI endpoints
## Summary
This PR refactors the Codex CLI authentication flow so that
**non-OpenAI** providers (for example **azure**, or any future addition)
can supply their API key through a dedicated environment variable
without triggering the OpenAI login flow.
Key behaviours introduced:
* When `provider !== "openai"` the CLI consults `src/utils/providers.ts`
to locate the correct environment variable (`AZURE_OPENAI_API_KEY`,
`GEMINI_API_KEY`, and so on) before considering any interactive login.
* Credit redemption (`--free`) and PKCE login now run **only** when the
provider is OpenAI, eliminating unwanted browser prompts for Azure and
others.
* User-facing error messages are revamped to guide Azure users to
**[https://ai.azure.com/](https://ai.azure.com)** and show the exact
variable name they must set.
* All code paths still export `OPENAI_API_KEY` so legacy scripts
continue to operate unchanged.
---
## Example `config.json`
```jsonc
{
"model": "codex-mini",
"provider": "azure",
"providers": {
"azure": {
"name": "AzureOpenAI",
"baseURL": "https://ai-<project-name>.openai.azure.com/openai",
"envKey": "AZURE_OPENAI_API_KEY"
}
},
"history": {
"maxSize": 1000,
"saveHistory": true,
"sensitivePatterns": []
}
}
```
With this file in `~/.codex/config.json`, a single command line is
enough:
```bash
export AZURE_OPENAI_API_KEY="<your-key>"
codex "Hello from Azure"
```
No browser window opens, and the CLI works in entirely non-interactive
mode.
---
## Rationale
The new flow enables Codex to run **asynchronously** in sandboxed
environments such as GitHub Actions pipelines. By passing `--provider
azure` (or setting it in `config.json`) and exporting the correct key,
CI/CD jobs can invoke Codex without any ChatGPT-style login or PKCE
round-trip. This unlocks fully automated testing and deployment
scenarios.
---
## What’s changed
| File | Type | Description |
| ------------------------ | ------------------- |
-----------------------------------------------------------------------------------------------------------------------------
|
| `codex-cli/src/cli.tsx` | **feat / refactor** | +43 / -20 lines.
Imports `providers`, adds early provider-specific key lookup, gates
`--free` redemption, rewrites help text. |
| `src/utils/providers.ts` | **chore** | Now consumed by CLI for env-var
discovery. |
---
## How to test
```bash
# Azure example
export AZURE_OPENAI_API_KEY="<your-key>"
codex --provider azure "Automated run in CI"
# OpenAI example (unchanged behaviour)
codex --provider openai --login "Standard OpenAI flow"
```
Expected outcomes:
* Azure and other provider paths are non-interactive when provider flag
is passed.
* The CLI always sets `OPENAI_API_KEY` for backward compatibility.
---
## Checklist
* [x] Logic behind provider-specific env-var lookup added.
* [x] Redundant OpenAI login steps removed for other providers.
* [x] Unit tests cover new branches.
* [x] README and sample config updated.
* [x] CI passes on all supported Node versions.
---
**Related work**
* #92
* #769
* #1321
I have read the CLA Document and I hereby sign the CLA.
I noticed that `/clear` wasn't fully clearing chat history; it would
clear the chat history widgets _in the UI_, but the LLM still had access
to information from previous messages.
This PR renames `/clear` to `/new` for clarity as per Michael's
suggestion, resetting `app_state` to a fresh `ChatWidget`.
Now that we have published a GitHub Release that contains arm64 musl
artifacts for Linux, update the following scripts to take advantage of
them:
- `dotslash-config.json` now uses musl artifacts for the `linux-aarch64`
target
- `install_native_deps.sh` for the TypeScript CLI now includes
`codex-linux-sandbox-aarch64-unknown-linux-musl` instead of
`codex-linux-sandbox-aarch64-unknown-linux-gnu` for sandboxing
- `codex-cli/bin/codex.js` now checks for `aarch64-unknown-linux-musl`
artifacts instead of `aarch64-unknown-linux-gnu` ones
Users were running into issues with glibc mismatches on arm64 linux. In
the past, we did not provide a musl build for arm64 Linux because we had
trouble getting the openssl dependency to build correctly. Though today
I just tried the same trick in `Cargo.toml` that we were doing for
`x86_64-unknown-linux-musl` (using `openssl-sys` with `features =
["vendored"]`), so I'm not sure what problem we had in the past the
builds "just worked" today!
Though one tweak that did have to be made is that the integration tests
for Seccomp/Landlock empirically require longer timeouts on arm64 linux,
or at least on the `ubuntu-24.04-arm` GitHub Runner. As such, we change
the timeouts for arm64 in `codex-rs/linux-sandbox/tests/landlock.rs`.
Though in solving this problem, I decided I needed a turnkey solution
for testing the Linux build(s) from my Mac laptop, so this PR introduces
`.devcontainer/Dockerfile` and `.devcontainer/devcontainer.json` to
facilitate this. Detailed instructions are in `.devcontainer/README.md`.
We will update `dotslash-config.json` and other release-related scripts
in a follow-up PR.
This does not implement the full Login with ChatGPT experience, but it
should unblock people.
**What works**
* The `codex` multitool now has a `login` subcommand, so you can run
`codex login`, which should write `CODEX_HOME/auth.json` if you complete
the flow successfully. The TUI will now read the `OPENAI_API_KEY` from
`auth.json`.
* The TUI should refresh the token if it has expired and the necessary
information is in `auth.json`.
* There is a `LoginScreen` in the TUI that tells you to run `codex
login` if both (1) your model provider expects to use `OPENAI_API_KEY`
as its env var, and (2) `OPENAI_API_KEY` is not set.
**What does not work**
* The `LoginScreen` does not support the login flow from within the TUI.
Instead, it tells you to quit, run `codex login`, and then run `codex`
again.
* `codex exec` does read from `auth.json` yet, nor does it direct the
user to go through the login flow if `OPENAI_API_KEY` is not be found.
* The `maybeRedeemCredits()` function from `get-api-key.tsx` has not
been ported from TypeScript to `login_with_chatgpt.py` yet:
a67a67f325/codex-cli/src/utils/get-api-key.tsx (L84-L89)
**Implementation**
Currently, the OAuth flow requires running a local webserver on
`127.0.0.1:1455`. It seemed wasteful to incur the additional binary cost
of a webserver dependency in the Rust CLI just to support login, so
instead we implement this logic in Python, as Python has a `http.server`
module as part of its standard library. Specifically, we bundle the
contents of a single Python file as a string in the Rust CLI and then
use it to spawn a subprocess as `python3 -c
{{SOURCE_FOR_PYTHON_SERVER}}`.
As such, the most significant files in this PR are:
```
codex-rs/login/src/login_with_chatgpt.py
codex-rs/login/src/lib.rs
```
Now that the CLI may load `OPENAI_API_KEY` from the environment _or_
`CODEX_HOME/auth.json`, we need a new abstraction for reading/writing
this variable, so we introduce:
```
codex-rs/core/src/openai_api_key.rs
```
Note that `std::env::set_var()` is [rightfully] `unsafe` in Rust 2024,
so we use a LazyLock<RwLock<Option<String>>> to store `OPENAI_API_KEY`
so it is read in a thread-safe manner.
Ultimately, it should be possible to go through the entire login flow
from the TUI. This PR introduces a placeholder `LoginScreen` UI for that
right now, though the new `codex login` subcommand introduced in this PR
should be a viable workaround until the UI is ready.
**Testing**
Because the login flow is currently implemented in a standalone Python
file, you can test it without building any Rust code as follows:
```
rm -rf /tmp/codex_home && mkdir /tmp/codex_home
CODEX_HOME=/tmp/codex_home python3 codex-rs/login/src/login_with_chatgpt.py
```
For reference:
* the original TypeScript implementation was introduced in
https://github.com/openai/codex/pull/963
* support for redeeming credits was later added in
https://github.com/openai/codex/pull/974
This PR overhauls how active tool calls and completed tool calls are
displayed:
1. More use of colour to indicate success/failure and distinguish
between components like tool name+arguments
2. Previously, the entire `CallToolResult` was serialized to JSON and
pretty-printed. Now, we extract each individual `CallToolResultContent`
and print those
1. The previous solution was wasting space by unnecessarily showing
details of the `CallToolResult` struct to users, without formatting the
actual tool call results nicely
2. We're now able to show users more information from tool results in
less space, with nicer formatting when tools return JSON results
### Before:
<img width="1251" alt="Screenshot 2025-06-03 at 11 24 26"
src="https://github.com/user-attachments/assets/5a58f222-219c-4c53-ace7-d887194e30cf"
/>
### After:
<img width="1265" alt="image"
src="https://github.com/user-attachments/assets/99fe54d0-9ebe-406a-855b-7aa529b91274"
/>
## Future Work
1. Integrate image tool result handling better. We should be able to
display images even if they're not the first `CallToolResultContent`
2. Users should have some way to view the full version of truncated tool
results
3. It would be nice to add some left padding for tool results, make it
more clear that they are results. This is doable, just a little fiddly
due to the way `first_visible_line` scrolling works
4. There's almost certainly a better way to format JSON than "all on 1
line with spaces to make Ratatui wrapping work". But I think that works
OK for now.
This fixes a longstanding error in the Rust CLI where `codex.rs`
contained an errant `is_first_turn` check that would exclude the user
instructions for subsequent "turns" of a conversation when using the
responses API (i.e., when `previous_response_id` existed).
While here, renames `Prompt.instructions` to `Prompt.user_instructions`
since we now have quite a few levels of instructions floating around.
Also removed an unnecessary use of `clone()` in
`Prompt.get_full_instructions()`.
As explained in detail in the doc comment for `ParseMode::Lenient`, we
have observed that GPT-4.1 does not always generate a valid invocation
of `apply_patch`. Fortunately, the error is predictable, so we introduce
some new logic to the `codex-apply-patch` crate to recover from this
error.
Because we would like to avoid this becoming a de facto standard (as it
would be incompatible if `apply_patch` were provided as an actual
executable, unless we also introduced the lenient behavior in the
executable, as well), we require passing `ParseMode::Lenient` to
`parse_patch_text()` to make it clear that the caller is opting into
supporting this special case.
Note the analogous change to the TypeScript CLI was
https://github.com/openai/codex/pull/930. In addition to changing the
accepted input to `apply_patch`, it also introduced additional
instructions for the model, which we include in this PR.
Note that `apply-patch` does not depend on either `regex` or
`regex-lite`, so some of the checks are slightly more verbose to avoid
introducing this dependency.
That said, this PR does not leverage the existing
`extract_heredoc_body_from_apply_patch_command()`, which depends on
`tree-sitter` and `tree-sitter-bash`:
5a5aa89914/codex-rs/apply-patch/src/lib.rs (L191-L246)
though perhaps it should.
Previous to this PR, we always set `reasoning` when making a request
using the Responses API:
d7245cbbc9/codex-rs/core/src/client.rs (L108-L111)
Though if you tried to use the Rust CLI with `--model gpt-4.1`, this
would fail with:
```shell
"Unsupported parameter: 'reasoning.effort' is not supported with this model."
```
We take a cue from the TypeScript CLI, which does a check on the model
name:
d7245cbbc9/codex-cli/src/utils/agent/agent-loop.ts (L786-L789)
This PR does a similar check, though also adds support for the following
config options:
```
model_reasoning_effort = "low" | "medium" | "high" | "none"
model_reasoning_summary = "auto" | "concise" | "detailed" | "none"
```
This way, if you have a model whose name happens to start with `"o"` (or
`"codex"`?), you can set these to `"none"` to explicitly disable
reasoning, if necessary. (That said, it seems unlikely anyone would use
the Responses API with non-OpenAI models, but we provide an escape
hatch, anyway.)
This PR also updates both the TUI and `codex exec` to show `reasoning
effort` and `reasoning summaries` in the header.
Prior to this PR, there were two big misses in `chat_completions.rs`:
1. The loop in `stream_chat_completions()` was only including items of
type `ResponseItem::Message` when building up the `"messages"` JSON for
the `POST` request to the `chat/completions` endpoint. This fixes things
by ensuring other variants (`FunctionCall`, `LocalShellCall`, and
`FunctionCallOutput`) are included, as well.
2. In `process_chat_sse()`, we were not recording tool calls and were
only emitting items of type
`ResponseEvent::OutputItemDone(ResponseItem::Message)` to the stream.
Now we introduce `FunctionCallState`, which is used to accumulate the
`delta`s of type `tool_calls`, so we can ultimately emit a
`ResponseItem::FunctionCall`, when appropriate.
While function calling now appears to work for chat completions with my
local testing, I believe that there are still edge cases that are not
covered and that this codepath would benefit from a battery of
integration tests. (As part of that further cleanup, we should also work
to support streaming responses in the UI.)
The other important part of this PR is some cleanup in
`core/src/codex.rs`. In particular, it was hard to reason about how
`run_task()` was building up the list of messages to include in a
request across the various cases:
- Responses API
- Chat Completions API
- Responses API used in concert with ZDR
I like to think things are a bit cleaner now where:
- `zdr_transcript` (if present) contains all messages in the history of
the conversation, which includes function call outputs that have not
been sent back to the model yet
- `pending_input` includes any messages the user has submitted while the
turn is in flight that need to be injected as part of the next `POST` to
the model
- `input_for_next_turn` includes the tool call outputs that have not
been sent back to the model yet
This PR introduces a `hide_agent_reasoning` config option (that defaults
to `false`) that users can enable to make the output less verbose by
suppressing reasoning output.
To test, verified that this includes agent reasoning in the output:
```
echo hello | just exec
```
whereas this does not:
```
echo hello | just exec --config hide_agent_reasoning=false
```
This required changing `ts_println!()` to take `$self:ident`, which is a
bit more verbose, but the usability improvement seems worth it.
Also eliminated an unnecessary `.to_string()` while here.
