## Summary
- Prioritize provider-specific API keys over default Codex auth when
building requests
- Add test to ensure provider env var auth overrides default auth
## Testing
- `just fmt`
- `just fix` *(fails: `let` expressions in this position are unstable)*
- `cargo test --all-features` *(fails: `let` expressions in this
position are unstable)*
------
https://chatgpt.com/codex/tasks/task_i_68926a104f7483208f2c8fd36763e0e3
The docs and code do not match. It turns out the docs are "right" in
they are what we have been meaning to support, so this PR updates the
code:
ae88b69b09/README.md (L298-L302)
Support for `instructions.md` is a holdover from the TypeScript CLI, so
we are just going to drop support for it altogether rather than maintain
it in perpetuity.
## Summary
Forgot to remove this in #1869 last night! Too much of a performance hit
on the main thread. We can bring it back via an async thread on startup.
## Summary
Includes a new user message in the api payload which provides useful
environment context for the model, so it knows about things like the
current working directory and the sandbox.
## Testing
Updated unit tests
## Summary
Have seen these tests flaking over the course of today on different
boxes. `wiremock` seems to be generally written with tokio/threads in
mind but based on the weird panics from the tests, let's see if this
helps.
- Added a `/status` command, which will be useful when we update the
home screen to print less status.
- Moved `create_config_summary_entries` to common since it's used in a
few places.
- Noticed we inconsistently had periods in slash command descriptions
and just removed them everywhere.
- Noticed the diff description was overflowing so made it shorter.
This PR attempts to break `codex-rust-review.md` into sections so that
it is easier to consume.
It also adds a healthy new section on "Assertions in Tests" that has
been on my mind for awhile.
This script attempts to verify that:
- You have no local, uncommitted changes.
- You are on `main`
- The commit you are on exists on `main` also exists on the origin
`https://github.com/openai/codex`, i.e., it is not just a commit you
have pushed to your local version of `main`
As part of this, try to print better error message if/when these
conditions are violated.
Hardcoding to `prerelease: true` is a holdover from before we had
migrated to the Rust CLI for releases and decided on how we were doing
version numbers.
To date, I have had to change the release status from "prerelease" to
"actual release" manually through the GitHub Releases web page. This is
a semi-serious problem because I've discovered that it messes up
Homebrew's automation if the version number _looks_ like a real release
but turns out to be a prerelease. The release potentially gets skipped
from being published on Homebrew, so it's important to set the value
correctly from the start.
I verified that `steps.release_name.outputs.name` does not include the
`rust-v` prefix from the tag name.
https://github.com/openai/codex/pull/1868 is a related fix that was in
flight simultaenously, but after talking to @easong-openai, this:
- logs instead of renders for `BackgroundEvent`
- logs for `TurnDiff`
- renders for `PatchApplyEnd`
## Summary
A split-up PR of #1763 , stacked on top of a tools refactor #1858 to
make the change clearer. From the previous summary:
> Let's try something new: tell the model about the sandbox, and let it
decide when it will need to break the sandbox. Some local testing
suggests that it works pretty well with zero iteration on the prompt!
## Testing
- [x] Added unit tests
- [x] Tested locally and it appears to work smoothly!
## Summary
In an effort to make tools easier to work with and more configurable,
I'm introducing `ToolConfig` and updating `Prompt` to take in a general
list of Tools. I think this is simpler and better for a few reasons:
- We can easily assemble tools from various sources (our own harness,
mcp servers, etc.) and we can consolidate the logic for constructing the
logic in one place that is separate from serialization.
- client.rs no longer needs arbitrary config values, it just takes in a
list of tools to serialize
A hefty portion of the PR is now updating our conversion of
`mcp_types::Tool` to `OpenAITool`, but considering that @bolinfest
accurately called this out as a TODO long ago, I think it's time we
tackled it.
## Testing
- [x] Experimented locally, no changes, as expected
- [x] Added additional unit tests
- [x] Responded to rust-review
Previous to this PR, `ShutdownComplete` was not being handled correctly
in `codex exec`, so it always ended up printing the following to stderr:
```
ERROR codex_exec: Error receiving event: InternalAgentDied
```
Because we were not breaking out of the loop for `ShutdownComplete`,
inevitably `codex.next_event()` would get called again and
`rx_event.recv()` would fail and the error would get mapped to
`InternalAgentDied`:
ea7d3f27bd/codex-rs/core/src/codex.rs (L190-L197)
For reference, https://github.com/openai/codex/pull/1647 introduced the
`ShutdownComplete` variant.
## Summary
Escalating out of sandbox is (almost always) not going to fix
long-running commands timing out - therefore we should just pass the
failure back to the model instead of asking the user to re-run a command
that took a long time anyway.
## Testing
- [x] Ran locally with a timeout and confirmed this worked as expected
This PR started as an investigation with the goal of eliminating the use
of `unsafe { std::env::set_var() }` in `ollama/src/client.rs`, as
setting environment variables in a multithreaded context is indeed
unsafe and these tests were observed to be flaky, as a result.
Though as I dug deeper into the issue, I discovered that the logic for
instantiating `OllamaClient` under test scenarios was not quite right.
In this PR, I aimed to:
- share more code between the two creation codepaths,
`try_from_oss_provider()` and `try_from_provider_with_base_url()`
- use the values from `Config` when setting up Ollama, as we have
various mechanisms for overriding config values, so we should be sure
that we are always using the ultimate `Config` for things such as the
`ModelProviderInfo` associated with the `oss` id
Once this was in place,
`OllamaClient::try_from_provider_with_base_url()` could be used in unit
tests for `OllamaClient` so it was possible to create a properly
configured client without having to set environment variables.
I ended up force-pushing https://github.com/openai/codex/pull/1848
because CI jobs were not being triggered after updating the PR on
GitHub, so this spelling error sneaked through.
This adds support for easily running Codex backed by a local Ollama
instance running our new open source models. See
https://github.com/openai/gpt-oss for details.
If you pass in `--oss` you'll be prompted to install/launch ollama, and
it will automatically download the 20b model and attempt to use it.
We'll likely want to expand this with some options later to make the
experience smoother for users who can't run the 20b or want to run the
120b.
Co-authored-by: Michael Bolin <mbolin@openai.com>
https://github.com/openai/codex/pull/1835 has some messed up history.
This adds support for streaming chat completions, which is useful for ollama. We should probably take a very skeptical eye to the code introduced in this PR.
---------
Co-authored-by: Ahmed Ibrahim <aibrahim@openai.com>
To date, we have a number of hardcoded OpenAI model slug checks spread
throughout the codebase, which makes it hard to audit the various
special cases for each model. To mitigate this issue, this PR introduces
the idea of a `ModelFamily` that has fields to represent the existing
special cases, such as `supports_reasoning_summaries` and
`uses_local_shell_tool`.
There is a `find_family_for_model()` function that maps the raw model
slug to a `ModelFamily`. This function hardcodes all the knowledge about
the special attributes for each model. This PR then replaces the
hardcoded model name checks with checks against a `ModelFamily`.
Note `ModelFamily` is now available as `Config::model_family`. We should
ultimately remove `Config::model` in favor of
`Config::model_family::slug`.
Stream models thoughts and responses instead of waiting for the whole
thing to come through. Very rough right now, but I'm making the risk call to push through.
## Summary
Our recent change in #1737 can sometimes lead to the model confusing
AGENTS.md context as part of the message. But a little prompting and
formatting can help fix this!
## Testing
- Ran locally with a few different prompts to verify the model
behaves well.
- Updated unit tests
Previously, `codex exec` was printing `Warning: no file to write last
message to` as a warning to stderr even though `--output-last-message`
was not specified, which is wrong. This fixes the code and changes
`handle_last_message()` so that it is only called when
`last_message_path` is `Some`.
Bumps [serde_json](https://github.com/serde-rs/json) from 1.0.141 to
1.0.142.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/serde-rs/json/releases">serde_json's
releases</a>.</em></p>
<blockquote>
<h2>v1.0.142</h2>
<ul>
<li>impl Default for &Value (<a
href="https://redirect.github.com/serde-rs/json/issues/1265">#1265</a>,
thanks <a
href="https://github.com/aatifsyed"><code>@aatifsyed</code></a>)</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="1731167cd5"><code>1731167</code></a>
Release 1.0.142</li>
<li><a
href="e51c81450a"><code>e51c814</code></a>
Touch up PR 1265</li>
<li><a
href="84abbdb613"><code>84abbdb</code></a>
Merge pull request <a
href="https://redirect.github.com/serde-rs/json/issues/1265">#1265</a>
from aatifsyed/master</li>
<li><a
href="9206cc0150"><code>9206cc0</code></a>
feat: impl Default for &Value</li>
<li>See full diff in <a
href="https://github.com/serde-rs/json/compare/v1.0.141...v1.0.142">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps [toml](https://github.com/toml-rs/toml) from 0.9.2 to 0.9.4.
<details>
<summary>Commits</summary>
<ul>
<li><a
href="2126e6af51"><code>2126e6a</code></a>
chore: Release</li>
<li><a
href="fa2100a888"><code>fa2100a</code></a>
docs: Update changelog</li>
<li><a
href="0c75bbd6f7"><code>0c75bbd</code></a>
feat(toml): Expose DeInteger/DeFloat as_str/radix (<a
href="https://redirect.github.com/toml-rs/toml/issues/1021">#1021</a>)</li>
<li><a
href="e3d64dff47"><code>e3d64df</code></a>
feat(toml): Expose DeFloat::as_str</li>
<li><a
href="ffdd211033"><code>ffdd211</code></a>
feat(toml): Expose DeInteger::as_str/radix</li>
<li><a
href="9e7adcc7fa"><code>9e7adcc</code></a>
docs(readme): Fix links to crates (<a
href="https://redirect.github.com/toml-rs/toml/issues/1020">#1020</a>)</li>
<li><a
href="73d04e20b5"><code>73d04e2</code></a>
docs(readme): Fix links to crates</li>
<li><a
href="da667e8a7d"><code>da667e8</code></a>
chore: Release</li>
<li><a
href="b1327fbe7c"><code>b1327fb</code></a>
docs: Update changelog</li>
<li><a
href="fb5346827e"><code>fb53468</code></a>
fix(toml): Don't enable std in toml_writer (<a
href="https://redirect.github.com/toml-rs/toml/issues/1019">#1019</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/toml-rs/toml/compare/toml-v0.9.2...toml-v0.9.4">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Added:
* C-m for newline (not sure if this is actually treated differently to
Enter, but tui-textarea handles it and it doesn't hurt)
* C-d to delete one char forwards (same as Del)
* A-bksp to delete backwards one word
* A-arrows to navigate by word
The existing prompt is really bad. As a low-hanging fruit, let's correct
the apply_patch instructions - this helps smaller models successfully
apply patches.
Allows users to set their experimental_instructions_file in configs.
For example the below enables experimental instructions when running
`codex -p foo`.
```
[profiles.foo]
experimental_instructions_file = "/Users/foo/.codex/prompt.md"
```
# Testing
- ✅ Running against a profile with experimental_instructions_file works.
- ✅ Running against a profile without experimental_instructions_file
works.
- ✅ Running against no profile with experimental_instructions_file
works.
- ✅ Running against no profile without experimental_instructions_file
works.
This lets us show an accumulating diff across all patches in a turn.
Refer to the docs for TurnDiffTracker for implementation details.
There are multiple ways this could have been done and this felt like the
right tradeoff between reliability and completeness:
*Pros*
* It will pick up all changes to files that the model touched including
if they prettier or another command that updates them.
* It will not pick up changes made by the user or other agents to files
it didn't modify.
*Cons*
* It will pick up changes that the user made to a file that the model
also touched
* It will not pick up changes to codegen or files that were not modified
with apply_patch
## Summary
Users frequently complain about re-approving commands that have failed
for non-sandbox reasons. We can't diagnose with complete accuracy which
errors happened because of a sandbox failure, but we can start to
eliminate some common simple cases.
This PR captures the most common case I've seen, which is a `command not
found` error.
## Testing
- [x] Added unit tests
- [x] Ran a few cases locally
This replaces tui-textarea with a custom textarea component.
Key differences:
1. wrapped lines
2. better unicode handling
3. uses the native terminal cursor
This should perhaps be spun out into its own separate crate at some
point, but for now it's convenient to have it in-tree.
The following test script fails in the codex sandbox:
```
import multiprocessing
from multiprocessing import Lock, Process
def f(lock):
with lock:
print("Lock acquired in child process")
if __name__ == '__main__':
lock = Lock()
p = Process(target=f, args=(lock,))
p.start()
p.join()
```
with
```
Traceback (most recent call last):
File "/Users/david.hao/code/codex/codex-rs/cli/test.py", line 9, in <module>
lock = Lock()
^^^^^^
File "/Users/david.hao/.local/share/uv/python/cpython-3.12.9-macos-aarch64-none/lib/python3.12/multiprocessing/context.py", line 68, in Lock
return Lock(ctx=self.get_context())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/david.hao/.local/share/uv/python/cpython-3.12.9-macos-aarch64-none/lib/python3.12/multiprocessing/synchronize.py", line 169, in __init__
SemLock.__init__(self, SEMAPHORE, 1, 1, ctx=ctx)
File "/Users/david.hao/.local/share/uv/python/cpython-3.12.9-macos-aarch64-none/lib/python3.12/multiprocessing/synchronize.py", line 57, in __init__
sl = self._semlock = _multiprocessing.SemLock(
^^^^^^^^^^^^^^^^^^^^^^^^^
PermissionError: [Errno 1] Operation not permitted
```
After reading, adding this line to the sandbox configs fixes things -
MacOS multiprocessing appears to use sem_lock(), which opens an IPC
which is considered a disk write even though no file is created. I
interrogated ChatGPT about whether it's okay to loosen, and my
impression after reading is that it is, although would appreciate a
close look
Breadcrumb: You can run `cargo run -- debug seatbelt --full-auto <cmd>`
to test the sandbox
To make `--full-auto` safer, this PR updates the Seatbelt policy so that
a `SandboxPolicy` with a `writable_root` that contains a `.git/`
_directory_ will make `.git/` _read-only_ (though as a follow-up, we
should also consider the case where `.git` is a _file_ with a `gitdir:
/path/to/actual/repo/.git` entry that should also be protected).
The two major changes in this PR:
- Updating `SandboxPolicy::get_writable_roots_with_cwd()` to return a
`Vec<WritableRoot>` instead of a `Vec<PathBuf>` where a `WritableRoot`
can specify a list of read-only subpaths.
- Updating `create_seatbelt_command_args()` to honor the read-only
subpaths in `WritableRoot`.
The logic to update the policy is a fairly straightforward update to
`create_seatbelt_command_args()`, but perhaps the more interesting part
of this PR is the introduction of an integration test in
`tests/sandbox.rs`. Leveraging the new API in #1785, we test
`SandboxPolicy` under various conditions, including ones where `$TMPDIR`
is not readable, which is critical for verifying the new behavior.
To ensure that Codex can run its own tests, e.g.:
```
just codex debug seatbelt --full-auto -- cargo test if_git_repo_is_writable_root_then_dot_git_folder_is_read_only
```
I had to introduce the use of `CODEX_SANDBOX=sandbox`, which is
comparable to how `CODEX_SANDBOX_NETWORK_DISABLED=1` was already being
used.
Adding a comparable change for Landlock will be done in a subsequent PR.
Introduce conversation.create handler (handle_create_conversation) and
wire it in MessageProcessor.
Stack:
Top: #1783
Bottom: #1784
---------
Co-authored-by: Gabriel Peal <gpeal@users.noreply.github.com>
Without this change, it is challenging to create integration tests to
verify that the folders not included in `writable_roots` in
`SandboxPolicy::WorkspaceWrite` are read-only because, by default,
`get_writable_roots_with_cwd()` includes `TMPDIR`, which is where most
integrationt
tests do their work.
This introduces a `use_exact_writable_roots` option to disable the
default
includes returned by `get_writable_roots_with_cwd()`.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1785).
* #1765
* __->__ #1785
## Summary
- stream command stdout as `ExecCommandStdout` events
- forward streamed stdout to clients and ignore in human output
processor
- adjust call sites for new streaming API
This fixes a bug in insert_history_lines where writing
`Line::From(vec!["A".bold(), "B".into()])` would write "B" as bold,
because "B" didn't explicitly subtract bold.
- MCP server: add send-user-message tool to send user input to a running
Codex session
- Added an integration tests for the happy and sad paths
Changes:
• Add tool definition and schema.
• Expose tool in capabilities.
• Route and handle tool requests with validation.
• Tests for success, bad UUID, and missing session.
follow‑ups
• Listen path not implemented yet; the tool is present but marked “don’t
use yet” in code comments.
• Session run flag reset: clear running_session_id_set appropriately
after turn completion/errors.
This is the third PR in a stack.
Stack:
Final: #1686
Intermediate: #1751
First: #1750
- Add operation to summarize the context so far.
- The operation runs a compact task that summarizes the context.
- The operation clear the previous context to free the context window
- The operation didn't use `run_task` to avoid corrupting the session
- Add /compact in the tui
https://github.com/user-attachments/assets/e06c24e5-dcfb-4806-934a-564d425a919c
- Expose mcp_protocol from mcp-server for reuse in tests and callers.
- In MessageProcessor, detect structured ToolCallRequestParams in
tools/call and forward to a new handler.
- Add handle_new_tool_calls scaffold (returns error for now).
- Test helper: add send_send_user_message_tool_call to McpProcess to
send ConversationSendMessage requests;
This is the second PR in a stack.
Stack:
Final: #1686
Intermediate: #1751
First: #1750
# Summary
- Align MCP server responses with mcp_types by emitting [CallToolResult,
RequestId] instead of an object.
Update send-message result to a tagged enum: Ok or Error { message }.
# Why
Protocol compliance with current MCP schema.
# Tests
- Updated assertions in mcp_protocol.rs for create/stream/send/list and
error cases.
This is the first PR in a stack.
Stack:
Final: #1686
Intermediate: #1751
First: #1750
This delays the call to insert_history_lines until a redraw is
happening. Crucially, the new lines are inserted _after the viewport is
resized_. This results in fewer stray blank lines below the viewport
when modals (e.g. user approval) are closed.
when we enabled KKP in https://github.com/openai/codex/pull/1743, we
started receiving keyup events, but didn't expect them anywhere in our
code. for now, just don't dispatch them at all.
At 550 lines, `exec.rs` was a bit large. In particular, I found it hard
to locate the Seatbelt-related code quickly without a file with
`seatbelt` in the name, so this refactors things so:
- `spawn_command_under_seatbelt()` and dependent code moves to a new
`seatbelt.rs` file
- `spawn_child_async()` and dependent code moves to a new `spawn.rs`
file
This is a follow-up to https://github.com/openai/codex/pull/1705, as
that PR inadvertently lost the logic where `PatchApplyBeginEvent` and
`PatchApplyEndEvent` events were sent when patches were auto-approved.
Though as part of this fix, I believe this also makes an important
safety fix to `assess_patch_safety()`, as there was a case that returned
`SandboxType::None`, which arguably is the thing we were trying to avoid
in #1705.
On a high level, we want there to be only one codepath where
`apply_patch` happens, which should be unified with the patch to run
`exec`, in general, so that sandboxing is applied consistently for both
cases.
Prior to this change, `apply_patch()` in `core` would either:
* exit early, delegating to `exec()` to shell out to `apply_patch` using
the appropriate sandbox
* proceed to run the logic for `apply_patch` in memory
549846b29a/codex-rs/core/src/apply_patch.rs (L61-L63)
In this implementation, only the latter would dispatch
`PatchApplyBeginEvent` and `PatchApplyEndEvent`, though the former would
dispatch `ExecCommandBeginEvent` and `ExecCommandEndEvent` for the
`apply_patch` call (or, more specifically, the `codex
--codex-run-as-apply-patch PATCH` call).
To unify things in this PR, we:
* Eliminate the back half of the `apply_patch()` function, and instead
have it also return with `DelegateToExec`, though we add an extra field
to the return value, `user_explicitly_approved_this_action`.
* In `codex.rs` where we process `DelegateToExec`, we use
`SandboxType::None` when `user_explicitly_approved_this_action` is
`true`. This means **we no longer run the apply_patch logic in memory**,
as we always `exec()`. (Note this is what allowed us to delete so much
code in `apply_patch.rs`.)
* In `codex.rs`, we further update `notify_exec_command_begin()` and
`notify_exec_command_end()` to take additional fields to determine what
type of notification to send: `ExecCommand` or `PatchApply`.
Admittedly, this PR also drops some of the functionality about giving
the user the opportunity to expand the set of writable roots as part of
approving the `apply_patch` command. I'm not sure how much that was
used, and we should probably rethink how that works as we are currently
tidying up the protocol to the TUI, in general.
this fixes a couple of panics that would happen when trying to render
something larger than the terminal, or insert history lines when the top
of the viewport is at y=0.
the git tests were failing on my local machine due to gpg signing config
in my ~/.gitconfig. tests should not be affected by ~/.gitconfig, so
configure them to ignore it.
Simplify and improve many UI elements.
* Remove all-around borders in most places. These interact badly with
terminal resizing and look heavy. Prefer left-side-only borders.
* Make the viewport adjust to the size of its contents.
* <kbd>/</kbd> and <kbd>@</kbd> autocomplete boxes appear below the
prompt, instead of above it.
* Restyle the keyboard shortcut hints & move them to the left.
* Restyle the approval dialog.
* Use synchronized rendering to avoid flashing during rerenders.
https://github.com/user-attachments/assets/96f044af-283b-411c-b7fc-5e6b8a433c20
<img width="1117" height="858" alt="Screenshot 2025-07-30 at 5 29 20 PM"
src="https://github.com/user-attachments/assets/0cc0af77-8396-429b-b6ee-9feaaccdbee7"
/>
The goal of this change is to try an experiment where we try to get AI
to take on more of the code review load. The idea is that once you
believe your PR is ready for review, please add the `codex-rust-review`
label (as opposed to the `codex-review` label).
Admittedly the corresponding prompt currently represents my personal
biases in terms of code review, but we should massage it over time to
represent the team's preferences.
Proof of concept for a resizable viewport.
The general approach here is to duplicate the `Terminal` struct from
ratatui, but with our own logic. This is a "light fork" in that we are
still using all the base ratatui functions (`Buffer`, `Widget` and so
on), but we're doing our own bookkeeping at the top level to determine
where to draw everything.
This approach could use improvement—e.g, when the window is resized to a
smaller size, if the UI wraps, we don't correctly clear out the
artifacts from wrapping. This is possible with a little work (i.e.
tracking what parts of our UI would have been wrapped), but this
behavior is at least at par with the existing behavior.
https://github.com/user-attachments/assets/4eb17689-09fd-4daa-8315-c7ebc654986d
cc @joshka who might have Thoughts™
Building on the work of https://github.com/openai/codex/pull/1702, this
changes how a shell call to `apply_patch` is handled.
Previously, a shell call to `apply_patch` was always handled in-process,
never leveraging a sandbox. To determine whether the `apply_patch`
operation could be auto-approved, the
`is_write_patch_constrained_to_writable_paths()` function would check if
all the paths listed in the paths were writable. If so, the agent would
apply the changes listed in the patch.
Unfortunately, this approach afforded a loophole: symlinks!
* For a soft link, we could fix this issue by tracing the link and
checking whether the target is in the set of writable paths, however...
* ...For a hard link, things are not as simple. We can run `stat FILE`
to see if the number of links is greater than 1, but then we would have
to do something potentially expensive like `find . -inum <inode_number>`
to find the other paths for `FILE`. Further, even if this worked, this
approach runs the risk of a
[TOCTOU](https://en.wikipedia.org/wiki/Time-of-check_to_time-of-use)
race condition, so it is not robust.
The solution, implemented in this PR, is to take the virtual execution
of the `apply_patch` CLI into an _actual_ execution using `codex
--codex-run-as-apply-patch PATCH`, which we can run under the sandbox
the user specified, just like any other `shell` call.
This, of course, assumes that the sandbox prevents writing through
symlinks as a mechanism to write to folders that are not in the writable
set configured by the sandbox. I verified this by testing the following
on both Mac and Linux:
```shell
#!/usr/bin/env bash
set -euo pipefail
# Can running a command in SANDBOX_DIR write a file in EXPLOIT_DIR?
# Codex is run in SANDBOX_DIR, so writes should be constrianed to this directory.
SANDBOX_DIR=$(mktemp -d -p "$HOME" sandboxtesttemp.XXXXXX)
# EXPLOIT_DIR is outside of SANDBOX_DIR, so let's see if we can write to it.
EXPLOIT_DIR=$(mktemp -d -p "$HOME" sandboxtesttemp.XXXXXX)
echo "SANDBOX_DIR: $SANDBOX_DIR"
echo "EXPLOIT_DIR: $EXPLOIT_DIR"
cleanup() {
# Only remove if it looks sane and still exists
[[ -n "${SANDBOX_DIR:-}" && -d "$SANDBOX_DIR" ]] && rm -rf -- "$SANDBOX_DIR"
[[ -n "${EXPLOIT_DIR:-}" && -d "$EXPLOIT_DIR" ]] && rm -rf -- "$EXPLOIT_DIR"
}
trap cleanup EXIT
echo "I am the original content" > "${EXPLOIT_DIR}/original.txt"
# Drop the -s to test hard links.
ln -s "${EXPLOIT_DIR}/original.txt" "${SANDBOX_DIR}/link-to-original.txt"
cat "${SANDBOX_DIR}/link-to-original.txt"
if [[ "$(uname)" == "Linux" ]]; then
SANDBOX_SUBCOMMAND=landlock
else
SANDBOX_SUBCOMMAND=seatbelt
fi
# Attempt the exploit
cd "${SANDBOX_DIR}"
codex debug "${SANDBOX_SUBCOMMAND}" bash -lc "echo pwned > ./link-to-original.txt" || true
cat "${EXPLOIT_DIR}/original.txt"
```
Admittedly, this change merits a proper integration test, but I think I
will have to do that in a follow-up PR.
Adds a `CodexAuth` type that encapsulates information about available
auth modes and logic for refreshing the token.
Changes `Responses` API to send requests to different endpoints based on
the auth type.
Updates login_with_chatgpt to support API-less mode and skip the key
exchange.
This adds a tool the model can call to update a plan. The tool doesn't
actually _do_ anything but it gives clients a chance to read and render
the structured plan. We will likely iterate on the prompt and tools
exposed for planning over time.
see
[discussion](https://github.com/rhysd/tui-textarea/issues/51#issuecomment-3021191712),
it's surprising that ^U behaves this way. IMO the undo/redo
functionality in tui-textarea isn't good enough to be worth preserving,
but if we do bring it back it should probably be on C-z / C-S-z / C-y.
Perhaps there was an intention to make the login screen prettier, but it
feels quite silly right now to just have a screen that says "press q",
so replace it with something that lets the user directly login without
having to quit the app.
<img width="1283" height="635" alt="Screenshot 2025-07-28 at 2 54 05 PM"
src="https://github.com/user-attachments/assets/f19e5595-6ef9-4a2d-b409-aa61b30d3628"
/>
## Summary
Per the [latest MCP
spec](https://modelcontextprotocol.io/specification/2025-06-18/basic#meta),
the `_meta` field is reserved for metadata. In the [Typescript
Schema](0695a497eb/schema/2025-06-18/schema.ts (L37-L40)),
`progressToken` is defined as a value to be attached to subsequent
notifications for that request.
The
[CallToolRequestParams](0695a497eb/schema/2025-06-18/schema.ts (L806-L817))
extends this definition but overwrites the params field. This ambiguity
makes our generated type definitions tricky, so I'm going to skip
`progressToken` field for now and just send back the `requestId`
instead.
In a future PR, we can clarify, update our `generate_mcp_types.py`
script, and update our progressToken logic accordingly.
## Testing
- [x] Added unit tests
- [x] Manually tested with mcp client
(Hopefully) temporary solution to the invisible approvals problem -
prints commands to history when they need approval and then also prints
the result of the approval. In the near future we should be able to do
some fancy stuff with updating commands before writing them to permanent
history.
Also, ctr-c while in the approval modal now acts as esc (aborts command)
and puts the TUI in the state where one additional ctr-c will exit.
This is a straight refactor, moving apply-patch-related code from
`codex.rs` and into the new `apply_patch.rs` file. The only "logical"
change is inlining `#[allow(clippy::unwrap_used)]` instead of declaring
`#![allow(clippy::unwrap_used)]` at the top of the file (which is
currently the case in `codex.rs`).
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1703).
* #1705
* __->__ #1703
* #1702
* #1698
* #1697
This introduces some special behavior to the CLIs that are using the
`codex-arg0` crate where if `arg1` is `--codex-run-as-apply-patch`, then
it will run as if `apply_patch arg2` were invoked. This is important
because it means we can do things like:
```
SANDBOX_TYPE=landlock # or seatbelt for macOS
codex debug "${SANDBOX_TYPE}" -- codex --codex-run-as-apply-patch PATCH
```
which gives us a way to run `apply_patch` while ensuring it adheres to
the sandbox the user specified.
While it would be nice to use the `arg0` trick like we are currently
doing for `codex-linux-sandbox`, there is no way to specify the `arg0`
for the underlying command when running under `/usr/bin/sandbox-exec`,
so it will not work for us in this case.
Admittedly, we could have also supported this via a custom environment
variable (e.g., `CODEX_ARG0`), but since environment variables are
inherited by child processes, that seemed like a potentially leakier
abstraction.
This change, as well as our existing reliance on checking `arg0`, place
additional requirements on those who include `codex-core`. Its
`README.md` has been updated to reflect this.
While we could have just added an `apply-patch` subcommand to the
`codex` multitool CLI, that would not be sufficient for the standalone
`codex-exec` CLI, which is something that we distribute as part of our
GitHub releases for those who know they will not be using the TUI and
therefore prefer to use a slightly smaller executable:
https://github.com/openai/codex/releases/tag/rust-v0.10.0
To that end, this PR adds an integration test to ensure that the
`--codex-run-as-apply-patch` option works with the standalone
`codex-exec` CLI.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1702).
* #1705
* #1703
* __->__ #1702
* #1698
* #1697
The overall idea here is: skip ratatui for writing into scrollback,
because its primitives are wrong. We want to render full lines of text,
that will be wrapped natively by the terminal, and which we never plan
to update using ratatui (so the `Buffer` struct is overhead and in fact
an inhibition).
Instead, we use ANSI scrolling regions (link reference doc to come).
Essentially, we:
1. Define a scrolling region that extends from the top of the prompt
area all the way to the top of scrollback
2. Scroll that region up by N < (screen_height - viewport_height) lines,
in this PR N=1
3. Put our cursor at the top of the newly empty region
4. Print out our new text like normal
The terminal interactions here (write_spans and its dependencies) are
mostly extracted from ratatui.
Most of the time, we expect the `String` returned by
`serde_json::to_string()` to have extra capacity, so `push('\n')` is
unlikely to allocate, which seems cheaper than an extra `write(2)` call,
on average?
This update replaces the previous ratatui history widget with an
append-only log so that the terminal can handle text selection and
scrolling. It also disables streaming responses, which we'll do our best
to bring back in a later PR. It also adds a small summary of token use
after the TUI exits.
Currently, codex on start shows the value for the approval policy as
name of
[AskForApproval](2437a8d17a/codex-rs/core/src/protocol.rs (L128))
enum, which differs from
[approval_policy](2437a8d17a/codex-rs/config.md (approval_policy))
config values.
E.g. "untrusted" becomes "UnlessTrusted", "on-failure" -> "OnFailure",
"never" -> "Never".
This PR changes render names of the approval policy to match with
configuration values.
This PR updates `is_known_safe_command()` to account for "safe
operators" to expand the set of commands that can be run without
approval. This concept existed in the TypeScript CLI, and we are
[finally!] porting it to the Rust one:
c9e2def494/codex-cli/src/approvals.ts (L531-L541)
The idea is that if we have `EXPR1 SAFE_OP EXPR2` and `EXPR1` and
`EXPR2` are considered safe independently, then `EXPR1 SAFE_OP EXPR2`
should be considered safe. Currently, `SAFE_OP` includes `&&`, `||`,
`;`, and `|`.
In the TypeScript implementation, we relied on
https://www.npmjs.com/package/shell-quote to parse the string of Bash,
as it could provide a "lightweight" parse tree, parsing `'beep || boop >
/byte'` as:
```
[ 'beep', { op: '||' }, 'boop', { op: '>' }, '/byte' ]
```
Though in this PR, we introduce the use of
https://crates.io/crates/tree-sitter-bash for parsing (which
incidentally we were already using in
[`codex-apply-patch`](c9e2def494/codex-rs/apply-patch/Cargo.toml (L18))),
which gives us a richer parse tree. (Incidentally, if you have never
played with tree-sitter, try the
[playground](https://tree-sitter.github.io/tree-sitter/7-playground.html)
and select **Bash** from the dropdown to see how it parses various
expressions.)
As a concrete example, prior to this change, our implementation of
`is_known_safe_command()` could verify things like:
```
["bash", "-lc", "grep -R \"Cargo.toml\" -n"]
```
but not:
```
["bash", "-lc", "grep -R \"Cargo.toml\" -n || true"]
```
With this change, the version with `|| true` is also accepted.
Admittedly, this PR does not expand the safety check to support
subshells, so it would reject, e.g. `bash -lc 'ls || (pwd && echo hi)'`,
but that can be addressed in a subsequent PR.
`nl` is a line-numbering tool that should be on the _trusted _ list, as
there is nothing concerning on https://gtfobins.github.io/gtfobins/nl/
that would merit exclusion.
`true` and `false` are also safe, though not particularly useful given
how `is_known_safe_command()` works today, but that will change with
https://github.com/openai/codex/pull/1668.
Because of a quirk of how implementation tests work in Rust, we had a
number of `#[allow(dead_code)]` annotations that were misleading because
the functions _were_ being used, just not by all integration tests in a
`tests/` folder, so when compiling the test that did not use the
function, clippy would complain that it was unused.
This fixes things by create a "test_support" crate under the `tests/`
folder that is imported as a dev dependency for the respective crate.
# Summary
- Writing effective evals for codex sessions requires context of the
overall repository state at the moment the session began
- This change adds this metadata (git repository, branch, commit hash)
to the top of the rollout of the session (if available - if not it
doesn't add anything)
- Currently, this is only effective on a clean working tree, as we can't
track uncommitted/untracked changes with the current metadata set.
Ideally in the future we may want to track unclean changes somehow, or
perhaps prompt the user to stash or commit them.
# Testing
- Added unit tests
- `cargo test && cargo clippy --tests && cargo fmt -- --config
imports_granularity=Item`
### Resulting Rollout
<img width="1243" height="127" alt="Screenshot 2025-07-17 at 1 50 00 PM"
src="https://github.com/user-attachments/assets/68108941-f015-45b2-985c-ea315ce05415"
/>
Bumps [toml](https://github.com/toml-rs/toml) from 0.9.1 to 0.9.2.
<details>
<summary>Commits</summary>
<ul>
<li><a
href="c28f9ac30f"><code>c28f9ac</code></a>
chore: Release</li>
<li><a
href="f3a2299148"><code>f3a2299</code></a>
docs: Update changelog</li>
<li><a
href="69f09d3093"><code>69f09d3</code></a>
fix(lex): Don't loop over ')' for forever (<a
href="https://redirect.github.com/toml-rs/toml/issues/1003">#1003</a>)</li>
<li><a
href="cc68ae4f42"><code>cc68ae4</code></a>
fix(lex): Don't loop over ')' for forever</li>
<li>See full diff in <a
href="https://github.com/toml-rs/toml/compare/toml-v0.9.1...toml-v0.9.2">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps [tree-sitter](https://github.com/tree-sitter/tree-sitter) from
0.25.6 to 0.25.8.
<details>
<summary>Commits</summary>
<ul>
<li><a
href="f2f197b6b2"><code>f2f197b</code></a>
0.25.8</li>
<li><a
href="8bb33f7d8c"><code>8bb33f7</code></a>
perf: reorder conditional operands</li>
<li><a
href="6f944de32f"><code>6f944de</code></a>
fix(generate): propagate node types error</li>
<li><a
href="c15938532d"><code>c159385</code></a>
0.25.7</li>
<li><a
href="94b55bfcdc"><code>94b55bf</code></a>
perf: reorder expensive conditional operand</li>
<li><a
href="bcb30f7951"><code>bcb30f7</code></a>
fix(generate): use topological sort for subtype map</li>
<li><a
href="3bd8f7df8e"><code>3bd8f7d</code></a>
perf: More efficient computation of used symbols</li>
<li><a
href="d7529c3265"><code>d7529c3</code></a>
perf: reserve <code>Vec</code> capacities where appropriate</li>
<li><a
href="bf4217f0ff"><code>bf4217f</code></a>
fix(web): wasm export paths</li>
<li><a
href="bb7b339ae2"><code>bb7b339</code></a>
Fix 'extra' field generation for node-types.json</li>
<li>Additional commits viewable in <a
href="https://github.com/tree-sitter/tree-sitter/compare/v0.25.6...v0.25.8">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps [rand](https://github.com/rust-random/rand) from 0.9.1 to 0.9.2.
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a
href="https://github.com/rust-random/rand/blob/master/CHANGELOG.md">rand's
changelog</a>.</em></p>
<blockquote>
<h2>[0.9.2 — 2025-07-20]</h2>
<h3>Deprecated</h3>
<ul>
<li>Deprecate <code>rand::rngs::mock</code> module and
<code>StepRng</code> generator (<a
href="https://redirect.github.com/rust-random/rand/issues/1634">#1634</a>)</li>
</ul>
<h3>Additions</h3>
<ul>
<li>Enable <code>WeightedIndex<usize></code> (de)serialization (<a
href="https://redirect.github.com/rust-random/rand/issues/1646">#1646</a>)</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="d3dd415705"><code>d3dd415</code></a>
Prepare rand_core 0.9.2 (<a
href="https://redirect.github.com/rust-random/rand/issues/1605">#1605</a>)</li>
<li><a
href="99fabd20e9"><code>99fabd2</code></a>
Prepare rand_core 0.9.2</li>
<li><a
href="c7fe1c43b5"><code>c7fe1c4</code></a>
rand: re-export <code>rand_core</code> (<a
href="https://redirect.github.com/rust-random/rand/issues/1604">#1604</a>)</li>
<li><a
href="db2b1e0bb4"><code>db2b1e0</code></a>
rand: re-export <code>rand_core</code></li>
<li><a
href="ee1d96f9f5"><code>ee1d96f</code></a>
rand_core: implement reborrow for <code>UnwrapMut</code> (<a
href="https://redirect.github.com/rust-random/rand/issues/1595">#1595</a>)</li>
<li><a
href="e0eb2ee0fc"><code>e0eb2ee</code></a>
rand_core: implement reborrow for <code>UnwrapMut</code></li>
<li><a
href="975f602f5d"><code>975f602</code></a>
fixup clippy 1.85 warnings</li>
<li><a
href="775b05be1b"><code>775b05b</code></a>
Relax <code>Sized</code> requirements for blanket impls (<a
href="https://redirect.github.com/rust-random/rand/issues/1593">#1593</a>)</li>
<li>See full diff in <a
href="https://github.com/rust-random/rand/compare/rand_core-0.9.1...rand_core-0.9.2">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
1. Emit call_id to exec approval elicitations for mcp client convenience
2. Remove the `-retry` from the call id for the same reason as above but
upstream the reset behavior to the mcp client
Always store the entire conversation history.
Request encrypted COT when not storing Responses.
Send entire input context instead of sending previous_response_id
This PR adds a `load_dotenv()` helper function to the `codex-common`
crate that is available when the `cli` feature is enabled. The function
uses [`dotenvy`](https://crates.io/crates/dotenvy) to update the
environment from:
- `$CODEX_HOME/.env`
- `$(pwd)/.env`
To test:
- ran `printenv OPENAI_API_KEY` to verify the env var exists in my
environment
- ran `just codex exec hello` to verify the CLI uses my `OPENAI_API_KEY`
- ran `unset OPENAI_API_KEY`
- ran `just codex exec hello` again and got **ERROR: Missing environment
variable: `OPENAI_API_KEY`**, as expected
- created `~/.codex/.env` and added `OPENAI_API_KEY=sk-proj-...` (also
ran `chmod 400 ~/.codex/.env` for good measure)
- ran `just codex exec hello` again and it worked, verifying it picked
up `OPENAI_API_KEY` from `~/.codex/.env`
Note this functionality was available in the TypeScript CLI:
https://github.com/openai/codex/pull/122 and was recently requested over
on https://github.com/openai/codex/issues/1262#issuecomment-3093203551.
I noticed that releases have taken longer and longer to build.
Originally, I think I did `--all-targets` to be confident that
everything builds cleanly, but that's really the job of CI that runs on
`main`, so we're spending a lot of time in `rust-release.yml` for not
that much additional signal.
Some users have reported issues where child processes are not cleaned up
after Codex exits (e.g., https://github.com/openai/codex/issues/1570).
This is generally a tricky issue on operating systems: if a parent
process receives `SIGKILL`, then it terminates immediately and cannot
communicate with the child.
**It only helps on Linux**, but this PR introduces the use of `prctl(2)`
so that if the parent process dies, `SIGTERM` will be delivered to the
child process. Whereas previously, I believe that if Codex spawned a
long-running process (like `tsc --watch`) and the Codex process received
`SIGKILL`, the `tsc --watch` process would be reparented to the init
process and would never be killed. Now with the use of `prctl(2)`, the
`tsc --watch` process should receive `SIGTERM` in that scenario.
We still need to come up with a solution for macOS. I've started to look
at `launchd`, but I'm researching a number of options.
1. Added an elicitation for `approve-patch` which is very similar to
`approve-exec`.
2. Extracted both elicitations to their own files to prevent
`codex_tool_runner` from blowing up in size.
## Summary
Adds a new mcp tool call, `codex-reply`, so we can continue existing
sessions. This is a first draft and does not yet support sessions from
previous processes.
## Testing
- [x] tested with mcp client
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`.
This updates the MCP server so that if it receives an
`ExecApprovalRequest` from the `Codex` session, it in turn sends an [MCP
elicitation](https://modelcontextprotocol.io/specification/draft/client/elicitation)
to the client to ask for the approval decision. Upon getting a response,
it forwards the client's decision via `Op::ExecApproval`.
Admittedly, we should be doing the same thing for
`ApplyPatchApprovalRequest`, but this is our first time experimenting
with elicitations, so I'm inclined to defer wiring that code path up
until we feel good about how this one works.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1623).
* __->__ #1623
* #1622
* #1621
* #1620
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
This updates the schema in `generate_mcp_types.py` from `2025-03-26` to
`2025-06-18`, regenerates `mcp-types/src/lib.rs`, and then updates all
the code that uses `mcp-types` to honor the changes.
Ran
```
npx @modelcontextprotocol/inspector just codex mcp
```
and verified that I was able to invoke the `codex` tool, as expected.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1621).
* #1623
* #1622
* __->__ #1621
## Summary
- extend rollout format to store all session data in JSON
- add resume/write helpers for rollouts
- track session state after each conversation
- support `LoadSession` op to resume a previous rollout
- allow starting Codex with an existing session via
`experimental_resume` config variable
We need a way later for exploring the available sessions in a user
friendly way.
## Testing
- `cargo test --no-run` *(fails: `cargo: command not found`)*
------
https://chatgpt.com/codex/tasks/task_i_68792a29dd5c832190bf6930d3466fba
This video is outdated. you should use `-c experimental_resume:<full
path>` instead of `--resume <full path>`
https://github.com/user-attachments/assets/7a9975c7-aa04-4f4e-899a-9e87defd947a
## Summary
- add OpenAI retry and timeout fields to Config
- inject these settings in tests instead of mutating env vars
- plumb Config values through client and chat completions logic
- document new configuration options
## Testing
- `cargo test -p codex-core --no-run`
------
https://chatgpt.com/codex/tasks/task_i_68792c5b04cc832195c03050c8b6ea94
---------
Co-authored-by: Michael Bolin <mbolin@openai.com>
This is designed to facilitate programmatic use of Codex in a more
lightweight way than using `codex mcp`.
Passing `--json` to `codex exec` will print each event as a line of JSON
to stdout. Note that it does not print the individual tokens as they are
streamed, only full messages, as this is aimed at programmatic use
rather than to power UI.
<img width="1348" height="1307" alt="image"
src="https://github.com/user-attachments/assets/fc7908de-b78d-46e4-a6ff-c85de28415c7"
/>
I changed the existing `EventProcessor` into a trait and moved the
implementation to `EventProcessorWithHumanOutput`. Then I introduced an
alternative implementation, `EventProcessorWithJsonOutput`. The `--json`
flag determines which implementation to use.
Adds a default vscode config with generally applicable settings.
Adds more entrypoints to justfile both for environment setup and to help
agents better verify changes.
When Codex CLI is installed via `npm`, we use a `.js` wrapper script to
launch the Rust binary.
- Previously, we were not listening for signals to ensure that killing
the Node.js process would also kill the underlying Rust process.
- We also did not have a proper `exit` handler in place on the child
process to ensure we exited from the Node.js process.
This PR fixes these things and hopefully addresses
https://github.com/openai/codex/issues/1570.
This also adds logic so that Windows falls back to the TypeScript CLI
again, which should address https://github.com/openai/codex/issues/1573.
This PR implements server name validation for MCP (Model Context
Protocol) servers to ensure they conform to the required pattern
^[a-zA-Z0-9_-]+$. This addresses the TODO comment in
mcp_connection_manager.rs:82.
+ Added validation before spawning MCP client tasks
+ Invalid server names are added to errors map with descriptive messages
I have read the CLA Document and I hereby sign the CLA
---------
Co-authored-by: Michael Bolin <bolinfest@gmail.com>
- Added support for message and reasoning deltas
- Skipped adding the support in the cli and tui for later
- Commented a failing test (wrong merge) that needs fix in a separate
PR.
Side note: I think we need to disable merge when the CI don't pass.
While this does make it so that `ctrl-d` will not exit Codex when the
composer is not empty, `ctrl-d` will still exit Codex if it is in the
"working" state.
Fixes https://github.com/openai/codex/issues/1443.
It appears that `0.5.0` was built with `stage_release.sh` instead of
`stage_rust_release.py`, so add docs to clarify this and recommend
running `--version` on the release candidate to verify the right thing
was built.
## Summary
- add integration test for chat mode streaming via CLI using wiremock
- add integration test for Responses API streaming via fixture
- call `cargo run` to invoke the CLI during tests
## Testing
- `cargo test -p codex-core --test cli_stream -- --nocapture`
- `cargo clippy --all-targets --all-features -- -D warnings`
------
https://chatgpt.com/codex/tasks/task_i_68715980bbec8321999534fdd6a013c1
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
In order to to this, I created a new `chatgpt` crate where we can put
any code that interacts directly with ChatGPT as opposed to the OpenAI
API. I added a disclaimer to the README for it that it should primarily
be modified by OpenAI employees.
https://github.com/user-attachments/assets/bb978e33-d2c9-4d8e-af28-c8c25b1988e8
https://github.com/openai/codex/pull/1524 introduced the new `config`
field on `ModelClient`, so this does the post-PR cleanup to remove the
now-unnecessary `model` field.
As noted in the updated docs, this makes it so that you can set:
```toml
model_supports_reasoning_summaries = true
```
as a way of overriding the existing heuristic for when to set the
`reasoning` field on a sampling request:
341c091c5b/codex-rs/core/src/client_common.rs (L152-L166)
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps node from 22-slim to 24-slim.
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps [toml](https://github.com/toml-rs/toml) from 0.9.0 to 0.9.1.
<details>
<summary>Commits</summary>
<ul>
<li><a
href="8c8ef44ea1"><code>8c8ef44</code></a>
chore: Release</li>
<li><a
href="b60ac5bfe9"><code>b60ac5b</code></a>
fix(toml): Correct minimal version for indexmap (<a
href="https://redirect.github.com/toml-rs/toml/issues/998">#998</a>)</li>
<li><a
href="966bd40511"><code>966bd40</code></a>
fix(toml): Correct minimal version for indexmap</li>
<li><a
href="2ed2af6519"><code>2ed2af6</code></a>
docs(readme): Mention additional crates</li>
<li>See full diff in <a
href="https://github.com/toml-rs/toml/compare/toml-v0.9.0...toml-v0.9.1">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
## Summary
Add Android platform support to Codex CLI
## What?
- Added `android` to the list of supported platforms in
`codex-cli/bin/codex.js`
- Treats Android as Linux for binary compatibility
## Why?
- Fixes "Unsupported platform: android (arm64)" error on Termux
- Enables Codex CLI usage on Android devices via Termux
- Improves platform compatibility without affecting other platforms
## How?
- Modified the platform detection switch statement to include `case
"android":`
- Android falls through to the same logic as Linux, using appropriate
ARM64 binaries
- Minimal change with no breaking effects on existing functionality
## Testing
- Tested on Android/Termux environment
- Verified the fix resolves the platform detection error
- Confirmed no impact on other platforms
## Related Issues
Fixes the "Unsupported platform: android (arm64)" error reported by
Termux users
Current 0.4.0 release:
```
~/code/codex2/codex-rs$ codex completion | head
_codex-cli() {
local i cur prev opts cmd
COMPREPLY=()
if [[ "${BASH_VERSINFO[0]}" -ge 4 ]]; then
cur="$2"
else
cur="${COMP_WORDS[COMP_CWORD]}"
fi
prev="$3"
cmd=""
```
with this change:
```
~/code/codex2/codex-rs$ just codex completion | head
cargo run --bin codex -- "$@"
Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.82s
Running `target/debug/codex completion`
_codex() {
local i cur prev opts cmd
COMPREPLY=()
if [[ "${BASH_VERSINFO[0]}" -ge 4 ]]; then
cur="$2"
else
cur="${COMP_WORDS[COMP_CWORD]}"
fi
prev="$3"
cmd=""
```
Some users have proxies or other setups where they are ultimately
hitting OpenAI endpoints, but need a custom `base_url` rather than the
default value of `"https://api.openai.com/v1"`. This PR makes it
possible to override the `base_url` for the `openai` provider via the
`OPENAI_BASE_URL` environment variable.
This is a stopgap solution before migrating the build for the npm
release to GitHub Actions (which is ultimately what should be done to
ensure hermetic builds).
The idea is that instead of continuing to create PRs like
https://github.com/openai/codex/pull/1472 where I have to check in a
change to the `WORKFLOW_URL`, this script uses `gh run list` to get the
`WORKFLOW_URL` dynamically and then threads the value through to
`install_native_deps.sh`.
To create the 0.3.0 release on npm, I ran:
```shell
./codex-cli/scripts/stage_rust_release.py --release-version 0.3.0
```
and then did `npm publish --dry-run` followed by `npm publish` in the
temp directory created by `stage_rust_release.py`.
On a high-level, we try to design `config.toml` so that you don't have
to "comment out a lot of stuff" when testing different options.
Previously, defining a sandbox policy was somewhat at odds with this
principle because you would define the policy as attributes of
`[sandbox]` like so:
```toml
[sandbox]
mode = "workspace-write"
writable_roots = [ "/tmp" ]
```
but if you wanted to temporarily change to a read-only sandbox, you
might feel compelled to modify your file to be:
```toml
[sandbox]
mode = "read-only"
# mode = "workspace-write"
# writable_roots = [ "/tmp" ]
```
Technically, commenting out `writable_roots` would not be strictly
necessary, as `mode = "read-only"` would ignore `writable_roots`, but
it's still a reasonable thing to do to keep things tidy.
Currently, the various values for `mode` do not support that many
attributes, so this is not that hard to maintain, but one could imagine
this becoming more complex in the future.
In this PR, we change Codex CLI so that it no longer recognizes
`[sandbox]`. Instead, it introduces a top-level option, `sandbox_mode`,
and `[sandbox_workspace_write]` is used to further configure the sandbox
when when `sandbox_mode = "workspace-write"` is used:
```toml
sandbox_mode = "workspace-write"
[sandbox_workspace_write]
writable_roots = [ "/tmp" ]
```
This feels a bit more future-proof in that it is less tedious to
configure different sandboxes:
```toml
sandbox_mode = "workspace-write"
[sandbox_read_only]
# read-only options here...
[sandbox_workspace_write]
writable_roots = [ "/tmp" ]
[sandbox_danger_full_access]
# danger-full-access options here...
```
In this scheme, you never need to comment out the configuration for an
individual sandbox type: you only need to redefine `sandbox_mode`.
Relatedly, previous to this change, a user had to do `-c
sandbox.mode=read-only` to change the mode on the command line. With
this change, things are arguably a bit cleaner because the equivalent
option is `-c sandbox_mode=read-only` (and now `-c
sandbox_workspace_write=...` can be set separately).
Though more importantly, we introduce the `-s/--sandbox` option to the
CLI, which maps directly to `sandbox_mode` in `config.toml`, making
config override behavior easier to reason about. Moreover, as you can
see in the updates to the various Markdown files, it is much easier to
explain how to configure sandboxing when things like `--sandbox
read-only` can be used as an example.
Relatedly, this cleanup also made it straightforward to add support for
a `sandbox` option for Codex when used as an MCP server (see the changes
to `mcp-server/src/codex_tool_config.rs`).
Fixes https://github.com/openai/codex/issues/1248.
v0.2.0 of https://www.npmjs.com/package/@openai/codex now runs the Rust
CLI, so it makes sense to bring back the instructions to use `npm i -g
@openai/codex`.
In most places, I list `npm install` before `brew install` because I
believe `npm` is more readily available, though I in the more detailed
part of the documentation, I note that `brew install` will download
fewer bytes, and in that sense, is preferred.
This adds support for two new model provider config options:
- `http_headers` for hardcoded (key, value) pairs
- `env_http_headers` for headers whose values should be read from
environment variables
This also updates the built-in `openai` provider to use this feature to
set the following headers:
- `originator` => `codex_cli_rs`
- `version` => [CLI version]
- `OpenAI-Organization` => `OPENAI_ORGANIZATION` env var
- `OpenAI-Project` => `OPENAI_PROJECT` env var
for consistency with the TypeScript implementation:
bd5a9e8ba9/codex-cli/src/utils/agent/agent-loop.ts (L321-L329)
While here, this also consolidates some logic that was duplicated across
`client.rs` and `chat_completions.rs` by introducing
`ModelProviderInfo.create_request_builder()`.
Resolves https://github.com/openai/codex/discussions/1152
This introduces two changes to make a quick fix so we can deploy the
Rust CLI for `0.2.0` of `@openai/codex` on npm:
- Updates `WORKFLOW_URL` to point to
https://github.com/openai/codex/actions/runs/15981617627, which is the
GitHub workflow run used to create the binaries for the `0.2.0` release
we published to Homebrew.
- Adds a `--version` option to `stage_release.sh` to specify what the
`version` field in the `package.json` will be.
Locally, I ran the following:
```
./codex-cli/scripts/stage_release.sh --native --version 0.2.0
```
Previously, we only used the `--native` flag to publish to the `native`
tag of `@openai/codex` (e.g., `npm publish --tag native`), but we should
just publish this as the default tag for `0.2.0` to be consistent with
what is in Homebrew.
We can still publish one "final" version of the TypeScript CLI as 0.1.x
later.
Under the hood, this release will still contain `dist/cli.js`,
`bin/codex-linux-sandbox-x64`, and `bin/codex-x86_64-apple-darwin`,
which are not strictly necessary, but we'll fix that in `0.3.0`.
As promised on https://github.com/openai/codex/discussions/1405, we are
making the first official release of the Rust CLI as v0.2.0. As part of
this move, we are making it available in Homebrew:
https://github.com/Homebrew/homebrew-core/pull/228615
Ultimately, we also plan to continue to make the CLI available in npm,
as well, though brew is a bit nicer in that `brew install` will download
only the binary for your platform whereas an npm module is expected to
contain the binaries for _all_ supported platforms, so it is a bit more
heavyweight.
A big part of this change is updating the root `README.md` to document
the behavior of the Rust CLI, which differs in a number of ways from the
TypeScript CLI. The existing `README.md` is moved to
`codex-cli/README.md` as part of this PR, as it is still applicable to
that folder.
As this is still early days for the Rust CLI, I encourage folks to
provide feedback on the command line flags and configuration options.
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.
Fixes:
* Instantiate `EventProcessor` earlier in `lib.rs` so
`print_config_summary()` can be an instance method of it and leverage
its various `Style` fields to ensure it honors `with_ansi` properly.
* After printing the config summary, print out user's prompt with the
heading `User instructions:`. As noted in the comment, now that we can
read the instructions via stdin as of #1178, it is helpful to the user
to ensure they know what instructions were given to Codex.
* Use same colors/bold/italic settings for headers as the TUI, making
the output a bit easier to read.
This attempts to make `codex exec` more flexible in how the prompt can
be passed:
* as before, it can be passed as a single string argument
* if `-` is passed as the value, the prompt is read from stdin
* if no argument is passed _and stdin is a tty_, prints a warning to
stderr that no prompt was specified an exits non-zero.
* if no argument is passed _and stdin is NOT a tty_, prints `Reading
prompt from stdin...` to stderr to let the user know that Codex will
wait until it reads EOF from stdin to proceed. (You can repro this case
by doing `yes | just exec` since stdin is not a TTY in that case but it
also never reaches EOF).
The main motivator behind this PR is that `stream_chat_completions()`
was not adding the `"tools"` entry to the payload posted to the
`/chat/completions` endpoint. This (1) refactors the existing logic to
build up the `"tools"` JSON from `client.rs` into `openai_tools.rs`, and
(2) updates the use of responses API (`client.rs`) and chat completions
API (`chat_completions.rs`) to both use it.
Note this PR alone is not sufficient to get tool calling from chat
completions working: that is done in
https://github.com/openai/codex/pull/1167.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/1177).
* #1167
* __->__ #1177
This is a first cut at a GitHub Action that lets you define prompt
templates in `.md` files under `.github/codex/labels` that will run
Codex with the associated prompt when the label is added to a GitHub
pull request.
For example, this PR includes these files:
```
.github/codex/labels/codex-attempt.md
.github/codex/labels/codex-code-review.md
.github/codex/labels/codex-investigate-issue.md
```
And the new `.github/workflows/codex.yml` workflow declares the
following triggers:
```yaml
on:
issues:
types: [opened, labeled]
pull_request:
branches: [main]
types: [labeled]
```
as well as the following expression to gate the action:
```
jobs:
codex:
if: |
(github.event_name == 'issues' && (
(github.event.action == 'labeled' && (github.event.label.name == 'codex-attempt' || github.event.label.name == 'codex-investigate-issue'))
)) ||
(github.event_name == 'pull_request' && github.event.action == 'labeled' && github.event.label.name == 'codex-code-review')
```
Note the "actor" who added the label must have write access to the repo
for the action to take effect.
After adding a label, the action will "ack" the request by replacing the
original label (e.g., `codex-review`) with an `-in-progress` suffix
(e.g., `codex-review-in-progress`). When it is finished, it will swap
the `-in-progress` label with a `-completed` one (e.g.,
`codex-review-completed`).
Users of the action are responsible for providing an `OPENAI_API_KEY`
and making it available as a secret to the action.
The way these definitions worked before, they did not handle quoted args
with spaces properly.
For example, if you had `/tmp/test-just/printlen.py` as:
```python
#!/usr/bin/env python3
import sys
print(len(sys.argv))
```
and your `justfile` was:
```
printlen *args:
/tmp/test-just/printlen.py {{args}}
```
Then:
```shell
$ just printlen foo bar
3
$ just printlen 'foo bar'
3
```
which is not what we want: `'foo bar'` should be treated as one
argument.
The fix is to use
[positional-arguments](515e806b51/README.md (L1131)):
```
set positional-arguments
printlen *args:
/tmp/test-just/printlen.py "$@"
```
The output of an MCP server tool call can be one of several types, but
to date, we treated all outputs as text by showing the serialized JSON
as the "tool output" in Codex:
25a9949c49/codex-rs/mcp-types/src/lib.rs (L96-L101)
This PR adds support for the `ImageContent` variant so we can now
display an image output from an MCP tool call.
In making this change, we introduce a new
`ResponseInputItem::McpToolCallOutput` variant so that we can work with
the `mcp_types::CallToolResult` directly when the function call is made
to an MCP server.
Though arguably the more significant change is the introduction of
`HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that
uses `ratatui_image` to render an image into the terminal. To support
this, we introduce `ImageRenderCache`, cache a
`ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the
appropriate scaled image data and dimensions based on the current
terminal size.
To test, I created a minimal `package.json`:
```json
{
"name": "kitty-mcp",
"version": "1.0.0",
"type": "module",
"description": "MCP that returns image of kitty",
"main": "index.js",
"dependencies": {
"@modelcontextprotocol/sdk": "^1.12.0"
}
}
```
with the following `index.js` to define the MCP server:
```js
#!/usr/bin/env node
import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import { readFile } from "node:fs/promises";
import { join } from "node:path";
const IMAGE_URI = "image://Ada.png";
const server = new McpServer({
name: "Demo",
version: "1.0.0",
});
server.tool(
"get-cat-image",
"If you need a cat image, this tool will provide one.",
async () => ({
content: [
{ type: "image", data: await getAdaPngBase64(), mimeType: "image/png" },
],
})
);
server.resource("Ada the Cat", IMAGE_URI, async (uri) => {
const base64Image = await getAdaPngBase64();
return {
contents: [
{
uri: uri.href,
mimeType: "image/png",
blob: base64Image,
},
],
};
});
async function getAdaPngBase64() {
const __dirname = new URL(".", import.meta.url).pathname;
// From 9705ce2c59/assets/Ada.png
const filePath = join(__dirname, "Ada.png");
const imageData = await readFile(filePath);
const base64Image = imageData.toString("base64");
return base64Image;
}
const transport = new StdioServerTransport();
await server.connect(transport);
```
With the local changes from this PR, I added the following to my
`config.toml`:
```toml
[mcp_servers.kitty]
command = "node"
args = ["/Users/mbolin/code/kitty-mcp/index.js"]
```
Running the TUI from source:
```
cargo run --bin codex -- --model o3 'I need a picture of a cat'
```
I get:
<img width="732" alt="image"
src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869"
/>
Now, that said, I have only tested in iTerm and there is definitely some
funny business with getting an accurate character-to-pixel ratio
(sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10
rows to render instead of 4), so there is still work to be done here.
The motivation behind this PR is to make it so a `HistoryCell` is more
like a `WidgetRef` that knows how to render itself into a `Rect` so that
it can be backed by something other than a `Vec<Line>`. Because a
`HistoryCell` is intended to appear in a scrollable list, we want to
ensure the stack of cells can be scrolled one `Line` at a time even if
the `HistoryCell` is not backed by a `Vec<Line>` itself.
To this end, we introduce the `CellWidget` trait whose key method is:
```
fn render_window(&self, first_visible_line: usize, area: Rect, buf: &mut Buffer);
```
The `first_visible_line` param is what differs from
`WidgetRef::render_ref()`, as a `CellWidget` needs to know the offset
into its "full view" at which it should start rendering.
The bookkeeping in `ConversationHistoryWidget` has been updated
accordingly to ensure each `CellWidget` in the history is rendered
appropriately.
This PR introduces support for `-c`/`--config` so users can override
individual config values on the command line using `--config
name=value`. Example:
```
codex --config model=o4-mini
```
Making it possible to set arbitrary config values on the command line
results in a more flexible configuration scheme and makes it easier to
provide single-line examples that can be copy-pasted from documentation.
Effectively, it means there are four levels of configuration for some
values:
- Default value (e.g., `model` currently defaults to `o4-mini`)
- Value in `config.toml` (e.g., user could override the default to be
`model = "o3"` in their `config.toml`)
- Specifying `-c` or `--config` to override `model` (e.g., user can
include `-c model=o3` in their list of args to Codex)
- If available, a config-specific flag can be used, which takes
precedence over `-c` (e.g., user can specify `--model o3` in their list
of args to Codex)
Now that it is possible to specify anything that could be configured in
`config.toml` on the command line using `-c`, we do not need to have a
custom flag for every possible config option (which can clutter the
output of `--help`). To that end, as part of this PR, we drop support
for the `--disable-response-storage` flag, as users can now specify `-c
disable_response_storage=true` to get the equivalent functionality.
Under the hood, this works by loading the `config.toml` into a
`toml::Value`. Then for each `key=value`, we create a small synthetic
TOML file with `value` so that we can run the TOML parser to get the
equivalent `toml::Value`. We then parse `key` to determine the point in
the original `toml::Value` to do the insert/replace. Once all of the
overrides from `-c` args have been applied, the `toml::Value` is
deserialized into a `ConfigToml` and then the `ConfigOverrides` are
applied, as before.
I discovered that if I ran `codex <PROMPT>` in a cwd that was not a Git
repo, Codex did not automatically run `<PROMPT>` after I accepted the
Git warning. It appears that we were not managing the `AppState`
transition correctly, so this fixes the bug and ensures the Codex
session does not start until the user accepts the Git warning.
In particular, we now create the `ChatWidget` lazily and store it in the
`AppState::Chat` variant.
Historically, we spawned the Seatbelt and Landlock sandboxes in
substantially different ways:
For **Seatbelt**, we would run `/usr/bin/sandbox-exec` with our policy
specified as an arg followed by the original command:
d1de7bb383/codex-rs/core/src/exec.rs (L147-L219)
For **Landlock/Seccomp**, we would do
`tokio::runtime::Builder::new_current_thread()`, _invoke
Landlock/Seccomp APIs to modify the permissions of that new thread_, and
then spawn the command:
d1de7bb383/codex-rs/core/src/exec_linux.rs (L28-L49)
While it is neat that Landlock/Seccomp supports applying a policy to
only one thread without having to apply it to the entire process, it
requires us to maintain two different codepaths and is a bit harder to
reason about. The tipping point was
https://github.com/openai/codex/pull/1061, in which we had to start
building up the `env` in an unexpected way for the existing
Landlock/Seccomp approach to continue to work.
This PR overhauls things so that we do similar things for Mac and Linux.
It turned out that we were already building our own "helper binary"
comparable to Mac's `sandbox-exec` as part of the `cli` crate:
d1de7bb383/codex-rs/cli/Cargo.toml (L10-L12)
We originally created this to build a small binary to include with the
Node.js version of the Codex CLI to provide support for Linux
sandboxing.
Though the sticky bit is that, at this point, we still want to deploy
the Rust version of Codex as a single, standalone binary rather than a
CLI and a supporting sandboxing binary. To satisfy this goal, we use
"the arg0 trick," in which we:
* use `std::env::current_exe()` to get the path to the CLI that is
currently running
* use the CLI as the `program` for the `Command`
* set `"codex-linux-sandbox"` as arg0 for the `Command`
A CLI that supports sandboxing should check arg0 at the start of the
program. If it is `"codex-linux-sandbox"`, it must invoke
`codex_linux_sandbox::run_main()`, which runs the CLI as if it were
`codex-linux-sandbox`. When acting as `codex-linux-sandbox`, we make the
appropriate Landlock/Seccomp API calls and then use `execvp(3)` to spawn
the original command, so do _replace_ the process rather than spawn a
subprocess. Incidentally, we do this before starting the Tokio runtime,
so the process should only have one thread when `execvp(3)` is called.
Because the `core` crate that needs to spawn the Linux sandboxing is not
a CLI in its own right, this means that every CLI that includes `core`
and relies on this behavior has to (1) implement it and (2) provide the
path to the sandboxing executable. While the path is almost always
`std::env::current_exe()`, we needed to make this configurable for
integration tests, so `Config` now has a `codex_linux_sandbox_exe:
Option<PathBuf>` property to facilitate threading this through,
introduced in https://github.com/openai/codex/pull/1089.
This common pattern is now captured in
`codex_linux_sandbox::run_with_sandbox()` and all of the `main.rs`
functions that should use it have been updated as part of this PR.
The `codex-linux-sandbox` crate added to the Cargo workspace as part of
this PR now has the bulk of the Landlock/Seccomp logic, which makes
`core` a bit simpler. Indeed, `core/src/exec_linux.rs` and
`core/src/landlock.rs` were removed/ported as part of this PR. I also
moved the unit tests for this code into an integration test,
`linux-sandbox/tests/landlock.rs`, in which I use
`env!("CARGO_BIN_EXE_codex-linux-sandbox")` as the value for
`codex_linux_sandbox_exe` since `std::env::current_exe()` is not
appropriate in that case.
https://github.com/openai/codex/pull/1086 is a work-in-progress to make
Linux sandboxing work more like Seatbelt where, for the command we want
to sandbox, we build up the command and then hand it, and some sandbox
configuration flags, to another command to set up the sandbox and then
run it.
In the case of Seatbelt, macOS provides this helper binary and provides
it at `/usr/bin/sandbox-exec`. For Linux, we have to build our own and
pass it through (which is what #1086 does), so this makes the new
`codex_linux_sandbox_exe` available on `Config` so that it will later be
available in `exec.rs` when we need it in #1086.
Added logic so that when we run `./scripts/stage_release.sh --native`
(for the `@native` version of the Node module), we drop a `use-native`
file next to `codex.js`. If present, `codex.js` will now run the Rust
CLI.
Ran `./scripts/stage_release.sh --native` and verified that when the
running `codex.js` in the staged folder:
```
$ /var/folders/wm/f209bc1n2bd_r0jncn9s6j_00000gp/T/tmp.efvEvBlSN6/bin/codex.js --version
codex-cli 0.0.2505220956
```
it ran the expected Rust version of the CLI, as desired.
While here, I also updated the Rust version to one that I cut today,
which includes the new shell environment policy config option:
https://github.com/openai/codex/pull/1061. Note this may "break" some
users if the processes spawned by Codex need extra environment
variables. (We are still working to determine what the right defaults
should be for this option.)
To date, when handling `shell` and `local_shell` tool calls, we were
spawning new processes using the environment inherited from the Codex
process itself. This means that the sensitive `OPENAI_API_KEY` that
Codex needs to talk to OpenAI models was made available to everything
run by `shell` and `local_shell`. While there are cases where that might
be useful, it does not seem like a good default.
This PR introduces a complex `shell_environment_policy` config option to
control the `env` used with these tool calls. It is inevitably a bit
complex so that it is possible to override individual components of the
policy so without having to restate the entire thing.
Details are in the updated `README.md` in this PR, but here is the
relevant bit that explains the individual fields of
`shell_environment_policy`:
| Field | Type | Default | Description |
| ------------------------- | -------------------------- | ------- |
-----------------------------------------------------------------------------------------------------------------------------------------------
|
| `inherit` | string | `core` | Starting template for the
environment:<br>`core` (`HOME`, `PATH`, `USER`, …), `all` (clone full
parent env), or `none` (start empty). |
| `ignore_default_excludes` | boolean | `false` | When `false`, Codex
removes any var whose **name** contains `KEY`, `SECRET`, or `TOKEN`
(case-insensitive) before other rules run. |
| `exclude` | array<string> | `[]` | Case-insensitive glob
patterns to drop after the default filter.<br>Examples: `"AWS_*"`,
`"AZURE_*"`. |
| `set` | table<string,string> | `{}` | Explicit key/value
overrides or additions – always win over inherited values. |
| `include_only` | array<string> | `[]` | If non-empty, a
whitelist of patterns; only variables that match _one_ pattern survive
the final step. (Generally used with `inherit = "all"`.) |
In particular, note that the default is `inherit = "core"`, so:
* if you have extra env variables that you want to inherit from the
parent process, use `inherit = "all"` and then specify `include_only`
* if you have extra env variables where you want to hardcode the values,
the default `inherit = "core"` will work fine, but then you need to
specify `set`
This configuration is not battle-tested, so we will probably still have
to play with it a bit. `core/src/exec_env.rs` has the critical business
logic as well as unit tests.
Though if nothing else, previous to this change:
```
$ cargo run --bin codex -- debug seatbelt -- printenv OPENAI_API_KEY
# ...prints OPENAI_API_KEY...
```
But after this change it does not print anything (as desired).
One final thing to call out about this PR is that the
`configure_command!` macro we use in `core/src/exec.rs` has to do some
complex logic with respect to how it builds up the `env` for the process
being spawned under Landlock/seccomp. Specifically, doing
`cmd.env_clear()` followed by `cmd.envs(&$env_map)` (which is arguably
the most intuitive way to do it) caused the Landlock unit tests to fail
because the processes spawned by the unit tests started failing in
unexpected ways! If we forgo `env_clear()` in favor of updating env vars
one at a time, the tests still pass. The comment in the code talks about
this a bit, and while I would like to investigate this more, I need to
move on for the moment, but I do plan to come back to it to fully
understand what is going on. For example, this suggests that we might
not be able to spawn a C program that calls `env_clear()`, which would
be...weird. We may still have to fiddle with our Landlock config if that
is the case.
Now the `exec` output starts with something like:
```
--------
workdir: /Users/mbolin/code/codex/codex-rs
model: o3
provider: openai
approval: Never
sandbox: SandboxPolicy { permissions: [DiskFullReadAccess, DiskWritePlatformUserTempFolder, DiskWritePlatformGlobalTempFolder, DiskWriteCwd, DiskWriteFolder { folder: "/Users/mbolin/.pyenv/shims" }] }
--------
```
which makes it easier to reason about when looking at logs.
`config.rs` is already quite long without these definitions. Since they
have no real dependencies of their own, let's move them to their own
file so `config.rs` can focus on the business logic of loading a config.
This introduces an experimental `--output-last-message` flag that can be
used to identify a file where the final message from the agent will be
written. Two use cases:
- Ultimately, we will likely add a `--quiet` option to `exec`, but even
if the user does not want any output written to the terminal, they
probably want to know what the agent did. Writing the output to a file
makes it possible to get that information in a clean way.
- Relatedly, when using `exec` in CI, it is easier to review the
transcript written "normally," (i.e., not as JSON or something with
extra escapes), but getting programmatic access to the last message is
likely helpful, so writing the last message to a file gets the best of
both worlds.
I am calling this "experimental" because it is possible that we are
overfitting and will want a more general solution to this problem that
would justify removing this flag.
2025-05-19 16:08:18 -07:00
271 changed files with 36850 additions and 6964 deletions
We provide the following options to facilitate Codex development in a container. This is particularly useful for verifying the Linux build when working on a macOS host.
## Docker
To build the Docker image locally for x64 and then run it with the repo mounted under `/workspace`:
Note that `/workspace/target` will contain the binaries built for your host platform, so we include `-e CARGO_TARGET_DIR=/workspace/codex-rs/target-amd64` in the `docker run` command so that the binaries built inside your container are written to a separate directory.
For arm64, specify `--platform=linux/amd64` instead for both `docker build` and `docker run`.
Currently, the `Dockerfile` works for both x64 and arm64 Linux, though you need to run `rustup target add x86_64-unknown-linux-musl` yourself to install the musl toolchain for x64.
## VS Code
VS Code recognizes the `devcontainer.json` file and gives you the option to develop Codex in a container. Currently, `devcontainer.json` builds and runs the `arm64` flavor of the container.
From the integrated terminal in VS Code, you can build either flavor of the `arm64` build (GNU or musl):
`openai/codex-action` is a GitHub Action that facilitates the use of [Codex](https://github.com/openai/codex) on GitHub issues and pull requests. Using the action, associate **labels** to run Codex with the appropriate prompt for the given context. Codex will respond by posting comments or creating PRs, whichever you specify!
Here is a sample workflow that uses `openai/codex-action`:
```yaml
name:Codex
on:
issues:
types:[opened, labeled]
pull_request:
branches:[main]
types:[labeled]
jobs:
codex:
if:...# optional, but can be effective in conserving CI resources
runs-on:ubuntu-latest
# TODO(mbolin): Need to verify if/when `write` is necessary.
permissions:
contents:write
issues:write
pull-requests:write
steps:
# By default, Codex runs network disabled using --full-auto, so perform
# any setup that requires network (such as installing dependencies)
See sample usage in [`codex.yml`](../../workflows/codex.yml).
## Triggering the Action
Using the sample workflow above, we have:
```yaml
on:
issues:
types:[opened, labeled]
pull_request:
branches:[main]
types:[labeled]
```
which means our workflow will be triggered when any of the following events occur:
- a label is added to an issue
- a label is added to a pull request against the `main` branch
### Label-Based Triggers
To define a GitHub label that should trigger Codex, create a file named `.github/codex/labels/LABEL-NAME.md` in your repository where `LABEL-NAME` is the name of the label. The content of the file is the prompt template to use when the label is added (see more on [Prompt Template Variables](#prompt-template-variables) below).
For example, if the file `.github/codex/labels/codex-review.md` exists, then:
- Adding the `codex-review` label will trigger the workflow containing the `openai/codex-action` GitHub Action.
- When `openai/codex-action` starts, it will replace the `codex-review` label with `codex-review-in-progress`.
- When `openai/codex-action` is finished, it will replace the `codex-review-in-progress` label with `codex-review-completed`.
If Codex sees that either `codex-review-in-progress` or `codex-review-completed` is already present, it will not perform the action.
As determined by the [default config](./src/default-label-config.ts), Codex will act on the following labels by default:
- Adding the `codex-review` label to a pull request will have Codex review the PR and add it to the PR as a comment.
- Adding the `codex-triage` label to an issue will have Codex investigate the issue and report its findings as a comment.
- Adding the `codex-issue-fix` label to an issue will have Codex attempt to fix the issue and create a PR wit the fix, if any.
## Action Inputs
The `openai/codex-action` GitHub Action takes the following inputs
### `openai_api_key` (required)
Set your `OPENAI_API_KEY` as a [repository secret](https://docs.github.com/en/actions/security-for-github-actions/security-guides/using-secrets-in-github-actions). See **Secrets and varaibles** then **Actions** in the settings for your GitHub repo.
Note that the secret name does not have to be `OPENAI_API_KEY`. For example, you might want to name it `CODEX_OPENAI_API_KEY` and then configure it on `openai/codex-action` as follows:
This is required so that Codex can post a comment or create a PR. Set this value on the action as follows:
```yaml
github_token:${{ secrets.GITHUB_TOKEN }}
```
### `codex_args`
A whitespace-delimited list of arguments to pass to Codex. Defaults to `--full-auto`, but if you want to override the default model to use `o3`:
```yaml
codex_args:"--full-auto --model o3"
```
For more complex configurations, use the `codex_home` input.
### `codex_home`
If set, the value to use for the `$CODEX_HOME` environment variable when running Codex. As explained [in the docs](https://github.com/openai/codex/tree/main/codex-rs#readme), this folder can contain the `config.toml` to configure Codex, custom instructions, and log files.
This should be a relative path within your repo.
## Prompt Template Variables
As shown above, `"prompt"` and `"promptPath"` are used to define prompt templates that will be populated and passed to Codex in response to certain events. All template variables are of the form `{CODEX_ACTION_...}` and the supported values are defined below.
### `CODEX_ACTION_ISSUE_TITLE`
If the action was triggered on a GitHub issue, this is the issue title.
Specifically it is read as the `.issue.title` from the `$GITHUB_EVENT_PATH`.
### `CODEX_ACTION_ISSUE_BODY`
If the action was triggered on a GitHub issue, this is the issue body.
Specifically it is read as the `.issue.body` from the `$GITHUB_EVENT_PATH`.
### `CODEX_ACTION_GITHUB_EVENT_PATH`
The value of the `$GITHUB_EVENT_PATH` environment variable, which is the path to the file that contains the JSON payload for the event that triggered the workflow. Codex can use `jq` to read only the fields of interest from this file.
### `CODEX_ACTION_PR_DIFF`
If the action was triggered on a pull request, this is the diff between the base and head commits of the PR. It is the output from `git diff`.
Note that the content of the diff could be quite large, so is generally safer to point Codex at `CODEX_ACTION_GITHUB_EVENT_PATH` and let it decide how it wants to explore the change.
description:"A reusable action that runs a Codex model."
inputs:
openai_api_key:
description:"The value to use as the OPENAI_API_KEY environment variable when running Codex."
required:true
trigger_phrase:
description:"Text to trigger Codex from a PR/issue body or comment."
required:false
default:""
github_token:
description:"Token so Codex can comment on the PR or issue."
required:true
codex_args:
description:"A whitespace-delimited list of arguments to pass to Codex. Due to limitations in YAML, arguments with spaces are not supported. For more complex configurations, use the `codex_home` input."
Provide a concise and respectful comment summarizing the findings.
### {CODEX_ACTION_ISSUE_TITLE}
{CODEX_ACTION_ISSUE_BODY}
`.trim(),
},
"codex-code-review":{
getPromptTemplate:()=>
`
Review this PR and respond with a very concise final message, formatted in Markdown.
There should be a summary of the changes (1-2 sentences) and a few bullet points if necessary.
Then provide the **review** (1-2 sentences plus bullet points, friendly tone).
{CODEX_ACTION_GITHUB_EVENT_PATH} contains the JSON that triggered this GitHub workflow. It contains the \`base\` and \`head\` refs that define this PR. Both refs are available locally.
`.trim(),
},
"codex-attempt-fix":{
getPromptTemplate:()=>
`
Attempt to solve the reported issue.
If a code change is required, create a new branch, commit the fix, and open a pull-request that resolves the problem.
Review this PR and respond with a very concise final message, formatted in Markdown.
There should be a summary of the changes (1-2 sentences) and a few bullet points if necessary.
Then provide the **review** (1-2 sentences plus bullet points, friendly tone).
{CODEX_ACTION_GITHUB_EVENT_PATH} contains the JSON that triggered this GitHub workflow. It contains the `base` and `head` refs that define this PR. Both refs are available locally.
Review this PR and respond with a very concise final message, formatted in Markdown.
There should be a summary of the changes (1-2 sentences) and a few bullet points if necessary.
Then provide the **review** (1-2 sentences plus bullet points, friendly tone).
Things to look out for when doing the review:
## General Principles
- **Make sure the pull request body explains the motivation behind the change.** If the author has failed to do this, call it out, and if you think you can deduce the motivation behind the change, propose copy.
- Ideally, the PR body also contains a small summary of the change. For small changes, the PR title may be sufficient.
- Each PR should ideally do one conceptual thing. For example, if a PR does a refactoring as well as introducing a new feature, push back and suggest the refactoring be done in a separate PR. This makes things easier for the reviewer, as refactoring changes can often be far-reaching, yet quick to review.
- When introducing new code, be on the lookout for code that duplicates existing code. When found, propose a way to refactor the existing code such that it should be reused.
## Code Organization
- Each create in the Cargo workspace in `codex-rs` has a specific purpose: make a note if you believe new code is not introduced in the correct crate.
- When possible, try to keep the `core` crate as small as possible. Non-core but shared logic is often a good candidate for `codex-rs/common`.
- Be wary of large files and offer suggestions for how to break things into more reasonably-sized files.
- Rust files should generally be organized such that the public parts of the API appear near the top of the file and helper functions go below. This is analagous to the "inverted pyramid" structure that is favored in journalism.
## Assertions in Tests
Assert the equality of the entire objects instead of doing "piecemeal comparisons," performing `assert_eq!()` on individual fields.
Note that unit tests also function as "executable documentation." As shown in the following example, "piecemeal comparisons" are often more verbose, provide less coverage, and are not as useful as executable documentation.
For example, suppose you have the following enum:
```rust
#[derive(Debug, PartialEq)]
enumMessage{
Request{
id: String,
method: String,
params: Option<serde_json::Value>,
},
Notification{
method: String,
params: Option<serde_json::Value>,
},
}
```
This is an example of a _piecemeal_ comparison:
```rust
// BAD: Piecemeal Comparison
#[test]
fntest_get_latest_messages(){
letmessages=get_latest_messages();
assert_eq!(messages.len(),2);
letm0=&messages[0];
matchm0{
Message::Request{id,method,params}=>{
assert_eq!(id,"123");
assert_eq!(method,"subscribe");
assert_eq!(
*params,
Some(json!({
"conversation_id": "x42z86"
}))
)
}
Message::Notification{..}=>{
panic!("expected Request");
}
}
letm1=&messages[1];
matchm1{
Message::Request{..}=>{
panic!("expected Notification");
}
Message::Notification{method,params}=>{
assert_eq!(method,"log");
assert_eq!(
*params,
Some(json!({
"level": "info",
"message": "subscribed"
}))
)
}
}
}
```
This is a _deep_ comparison:
```rust
// GOOD: Verify the entire structure with a single assert_eq!().
usepretty_assertions::assert_eq;
#[test]
fntest_get_latest_messages(){
letmessages=get_latest_messages();
assert_eq!(
vec![
Message::Request{
id: "123".to_string(),
method: "subscribe".to_string(),
params: Some(json!({
"conversation_id": "x42z86"
})),
},
Message::Notification{
method: "log".to_string(),
params: Some(json!({
"level": "info",
"message": "subscribed"
})),
},
],
messages,
);
}
```
## More Tactical Rust Things To Look Out For
- Do not use `unsafe` (unless you have a really, really good reason like using an operating system API directly and no safe wrapper exists). For example, there are cases where it is tempting to use `unsafe` in order to use `std::env::set_var()`, but this indeed `unsafe` and has led to race conditions on multiple occasions. (When this happens, find a mechanism other than environment variables to use for configuration.)
- Encourage the use of small enums or the newtype pattern in Rust if it helps readability without adding significant cognitive load or lines of code.
- If you see opportunities for the changes in a diff to use more idiomatic Rust, please make specific recommendations. For example, favor the use of expressions over `return`.
- When modifying a `Cargo.toml` file, make sure that dependency lists stay alphabetically sorted. Also consider whether a new dependency is added to the appropriate place (e.g., `[dependencies]` versus `[dev-dependencies]`)
## Pull Request Body
- If the nature of the change seems to have a visual component (which is often the case for changes to `codex-rs/tui`), recommend including a screenshot or video to demonstrate the change, if appropriate.
- References to existing GitHub issues and PRs are encouraged, where appropriate, though you likely do not have network access, so may not be able to help here.
# PR Information
{CODEX_ACTION_GITHUB_EVENT_PATH} contains the JSON that triggered this GitHub workflow. It contains the `base` and `head` refs that define this PR. Both refs are available locally.
- Never add or modify any code related to `CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR`. You operate in a sandbox where `CODEX_SANDBOX_NETWORK_DISABLED=1` will be set whenever you use the `shell` tool. Any existing code that uses `CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR` was authored with this fact in mind. It is often used to early exit out of tests that the author knew you would not be able to run given your sandbox limitations.
- Never add or modify any code related to `CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR` or `CODEX_SANDBOX_ENV_VAR`.
- You operate in a sandbox where `CODEX_SANDBOX_NETWORK_DISABLED=1` will be set whenever you use the `shell` tool. Any existing code that uses `CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR` was authored with this fact in mind. It is often used to early exit out of tests that the author knew you would not be able to run given your sandbox limitations.
- Similarly, when you spawn a process using Seatbelt (`/usr/bin/sandbox-exec`), `CODEX_SANDBOX=seatbelt` will be set on the child process. Integration tests that want to run Seatbelt themselves cannot be run under Seatbelt, so checks for `CODEX_SANDBOX=seatbelt` are also often used to early exit out of tests, as appropriate.
Before creating a pull request with changes to `codex-rs`, run `just fmt` (in `codex-rs` directory) to format the code and `just fix` (in `codex-rs` directory) to fix any linter issues in the code, ensure the test suite passes by running `cargo test --all-features` in the `codex-rs` directory.
When making individual changes prefer running tests on individual files or projects first.
<p align="center">Lightweight coding agent that runs in your terminal</p>
<p align="center"><code>npm i -g @openai/codex</code></p>
<p align="center"><code>npm i -g @openai/codex</code><br />or <code>brew install codex</code></p>

This is the home of the **Codex CLI**, which is a coding agent from OpenAI that runs locally on your computer. If you are looking for the _cloud-based agent_ from OpenAI, **Codex [Web]**, see <https://chatgpt.com/codex>.
<!--  -->
Or go to the [latest GitHub Release](https://github.com/openai/codex/releases/latest) and download the appropriate binary for your platform.
### OpenAI API Users
Next, set your OpenAI API key as an environment variable:
```shell
exportOPENAI_API_KEY="your-api-key-here"
```
> **Note:** This command sets the key only for your current terminal session. You can add the `export` line to your shell's configuration file (e.g., `~/.zshrc`) but we recommend setting for the session. **Tip:** You can also place your API key into a `.env` file at the root of your project:
>
> ```env
> OPENAI_API_KEY=your-api-key-here
> ```
>
> The CLI will automatically load variables from `.env` (via `dotenv/config`).
> [!NOTE]
>This command sets the key only for your current terminal session. You can add the `export` line to your shell's configuration file (e.g., `~/.zshrc`), but we recommend setting it for the session.
### OpenAI Plus/Pro Users
If you have a paid OpenAI account, run the following to start the login process:
```
codex login
```
If you complete the process successfully, you should have a `~/.codex/auth.json` file that contains the credentials that Codex will use.
To verify whether you are currently logged in, run:
```
codex login status
```
If you encounter problems with the login flow, please comment on <https://github.com/openai/codex/issues/1243>.
<details>
<summary><strong>Use <code>--provider</code> to use other models</strong></summary>
<summary><strong>Use <code>--profile</code> to use other models</strong></summary>
> Codex also allows you to use other providers that support the OpenAI Chat Completions API. You can set the provider in the config file or use the `--provider` flag. The possible options for `--provider` are:
>
> - openai (default)
> - openrouter
> - azure
> - gemini
> - ollama
> - mistral
> - deepseek
> - xai
> - groq
> - arceeai
> - any other provider that is compatible with the OpenAI API
>
> If you use a provider other than OpenAI, you will need to set the API key for the provider in the config file or in the environment variable as:
>
> ```shell
> export <provider>_API_KEY="your-api-key-here"
> ```
>
> If you use a provider not listed above, you must also set the base URL for the provider:
Codex also allows you to use other providers that support the OpenAI Chat Completions (or Responses) API.
To do so, you must first define custom [providers](./config.md#model_providers) in `~/.codex/config.toml`. For example, the provider for a standard Ollama setup would be defined as follows:
```toml
[model_providers.ollama]
name="Ollama"
base_url="http://localhost:11434/v1"
```
The `base_url` will have `/chat/completions` appended to it to build the full URL for the request.
For providers that also require an `Authorization` header of the form `Bearer: SECRET`, an `env_key` can be specified, which indicates the environment variable to read to use as the value of `SECRET` when making a request:
```toml
[model_providers.openrouter]
name="OpenRouter"
base_url="https://openrouter.ai/api/v1"
env_key="OPENROUTER_API_KEY"
```
Providers that speak the Responses API are also supported by adding `wire_api = "responses"` as part of the definition. Accessing OpenAI models via Azure is an example of such a provider, though it also requires specifying additional `query_params` that need to be appended to the request URL:
```toml
[model_providers.azure]
name="Azure"
# Make sure you set the appropriate subdomain for this URL.
env_key="AZURE_OPENAI_API_KEY"# Or "OPENAI_API_KEY", whichever you use.
# Newer versions appear to support the responses API, see https://github.com/openai/codex/pull/1321
query_params={api-version="2025-04-01-preview"}
wire_api="responses"
```
Once you have defined a provider you wish to use, you can configure it as your default provider as follows:
```toml
model_provider="azure"
```
> [!TIP]
> If you find yourself experimenting with a variety of models and providers, then you likely want to invest in defining a _profile_ for each configuration like so:
```toml
[profiles.o3]
model_provider="azure"
model="o3"
[profiles.mistral]
model_provider="ollama"
model="mistral"
```
This way, you can specify one command-line argument (.e.g., `--profile o3`, `--profile mistral`) to override multiple settings together.
</details>
<br />
@@ -136,7 +178,7 @@ codex "explain this codebase to me"
```
```shell
codex --approval-mode full-auto "create the fanciest todo-list app"
codex --full-auto "create the fanciest todo-list app"
```
That's it - Codex will scaffold a file, run it inside a sandbox, install any
@@ -145,6 +187,41 @@ they'll be committed to your working directory.
---
## Using Open Source Models
Codex can run fully locally against an OpenAI-compatible OSS host (like Ollama) using the `--oss` flag:
- Interactive UI:
- codex --oss
- Non-interactive (programmatic) mode:
- echo "Refactor utils" | codex exec --oss
Model selection when using `--oss`:
- If you omit `-m/--model`, Codex defaults to -m gpt-oss:20b and will verify it exists locally (downloading if needed).
- To pick a different size, pass one of:
- -m "gpt-oss:20b"
- -m "gpt-oss:120b"
Point Codex at your own OSS host:
- By default, `--oss` talks to http://localhost:11434/v1.
- To use a different host, set one of these environment variables before running Codex:
| **Full Auto** | <li>Read/write files <li> Execute shell commands (network disabled, writes limited to your workdir) | - |
- [`approval_policy`](./codex-rs/config.md#approval_policy) determines when you should be prompted to approve whether Codex can execute a command
- [`sandbox`](./codex-rs/config.md#sandbox) determines the _sandbox policy_ that Codex uses to execute untrusted commands
In **Full Auto** every command is run **network-disabled** and confined to the
current working directory (plus temporary files) for defense-in-depth. Codex
will also show a warning/confirmation if you start in **auto-edit** or
**full-auto** while the directory is _not_ tracked by Git, so you always have a
safety net.
By default, Codex runs with `--ask-for-approval untrusted` and `--sandbox read-only`, which means that:
Coming soon: you'll be able to whitelist specific commands to auto-execute with
the network enabled, once we're confident in additional safeguards.
- The user is prompted to approve every command not on the set of "trusted" commands built into Codex (`cat`, `ls`, etc.)
- Approved commands are run outside of a sandbox because user approval implies "trust," in this case.
Running Codex with the `--full-auto` convenience flag changes the configuration to `--ask-for-approval on-failure` and `--sandbox workspace-write`, which means that:
- Codex does not initially ask for user approval before running an individual command.
- Though when it runs a command, it is run under a sandbox in which:
- It can read any file on the system.
- It can only write files under the current directory (or the directory specified via `--cd`).
- Network requests are completely disabled.
- Only if the command exits with a non-zero exit code will it ask the user for approval. If granted, it will re-attempt the command outside of the sandbox. (A common case is when Codex cannot `npm install` a dependency because that requires network access.)
Again, these two options can be configured independently. For example, if you want Codex to perform an "exploration" where you are happy for it to read anything it wants but you never want to be prompted, you could run Codex with `--ask-for-approval never` and `--sandbox read-only`.
### Platform sandboxing details
The hardening mechanism Codex uses depends on your OS:
The mechanism Codex uses to implement the sandbox policy depends on your OS:
- **macOS 12+** - commands are wrapped with **Apple Seatbelt** (`sandbox-exec`).
- **macOS 12+** uses **Apple Seatbelt** and runs commands using `sandbox-exec` with a profile (`-p`) that corresponds to the `--sandbox` that was specified.
- **Linux** uses a combination of Landlock/seccomp APIs to enforce the `sandbox` configuration.
- Everything is placed in a read-only jail except for a small set of
- Outbound network is _fully blocked_ by default - even if a child process
tries to `curl` somewhere it will fail.
- **Linux** - there is no sandboxing by default.
We recommend using Docker for sandboxing, where Codex launches itself inside a **minimal
container image** and mounts your repo _read/write_ at the same path. A
custom `iptables`/`ipset` firewall script denies all egress except the
OpenAI API. This gives you deterministic, reproducible runs without needing
root on the host. You can use the [`run_in_container.sh`](./codex-cli/scripts/run_in_container.sh) script to set up the sandbox.
Note that when running Linux in a containerized environment such as Docker, sandboxing may not work if the host/container configuration does not support the necessary Landlock/seccomp APIs. In such cases, we recommend configuring your Docker container so that it provides the sandbox guarantees you are looking for and then running `codex` with `--sandbox danger-full-access` (or, more simply, the `--dangerously-bypass-approvals-and-sandbox` flag) within your container.
---
@@ -205,24 +276,20 @@ The hardening mechanism Codex uses depends on your OS:
Key flags: `--model/-m`, `--approval-mode/-a`, `--quiet/-q`, and `--notify`.
Key flags: `--model/-m`, `--ask-for-approval/-a`.
---
@@ -234,8 +301,6 @@ You can give Codex extra instructions and guidance using `AGENTS.md` files. Code
2.`AGENTS.md` at repo root - shared project notes
3.`AGENTS.md` in the current working directory - sub-folder/feature specifics
Disable loading of these files with `--no-project-doc` or the environment variable `CODEX_DISABLE_PROJECT_DOC=1`.
---
## Non-interactive / CI mode
@@ -247,18 +312,37 @@ Run Codex head-less in pipelines. Example GitHub Action step:
run:|
npm install -g @openai/codex
export OPENAI_API_KEY="${{ secrets.OPENAI_KEY }}"
codex -a auto-edit --quiet "update CHANGELOG for next release"
codex exec --full-auto "update CHANGELOG for next release"
```
Set `CODEX_QUIET_MODE=1` to silence interactive UI noise.
## Model Context Protocol (MCP)
The Codex CLI can be configured to leverage MCP servers by defining an [`mcp_servers`](./codex-rs/config.md#mcp_servers) section in `~/.codex/config.toml`. It is intended to mirror how tools such as Claude and Cursor define `mcpServers` in their respective JSON config files, though the Codex format is slightly different since it uses TOML rather than JSON, e.g.:
```toml
# IMPORTANT: the top-level key is `mcp_servers` rather than `mcpServers`.
[mcp_servers.server-name]
command="npx"
args=["-y","mcp-server"]
env={"API_KEY"="value"}
```
> [!TIP]
> It is somewhat experimental, but the Codex CLI can also be run as an MCP _server_ via `codex mcp`. If you launch it with an MCP client such as `npx @modelcontextprotocol/inspector codex mcp` and send it a `tools/list` request, you will see that there is only one tool, `codex`, that accepts a grab-bag of inputs, including a catch-all `config` map for anything you might want to override. Feel free to play around with it and provide feedback via GitHub issues.
## Tracing / verbose logging
Setting the environment variable `DEBUG=true` prints full API request and response details:
Because Codex is written in Rust, it honors the `RUST_LOG` environment variable to configure its logging behavior.
The TUI defaults to `RUST_LOG=codex_core=info,codex_tui=info` and log messages are written to `~/.codex/log/codex-tui.log`, so you can leave the following running in a separate terminal to monitor log messages as they are written:
```shell
DEBUG=true codex
```
tail -F ~/.codex/log/codex-tui.log
```
By comparison, the non-interactive mode (`codex exec`) defaults to `RUST_LOG=error`, but messages are printed inline, so there is no need to monitor a separate file.
See the Rust documentation on [`RUST_LOG`](https://docs.rs/env_logger/latest/env_logger/#enabling-logging) for more information on the configuration options.
---
@@ -281,201 +365,95 @@ Below are a few bite-size examples you can copy-paste. Replace the text in quote
The GitHub Release also contains a [DotSlash](https://dotslash-cli.com/) file for the Codex CLI named `codex`. Using a DotSlash file makes it possible to make a lightweight commit to source control to ensure all contributors use the same version of an executable, regardless of what platform they use for development.
</details>
<details>
<summary><strong>Build from source</strong></summary>
```bash
# Clone the repository and navigate to the CLI package
# Clone the repository and navigate to the root of the Cargo workspace.
Though `--config` can be used to set/override ad-hoc config values for individual invocations of `codex`.
---
@@ -524,7 +502,13 @@ Codex CLI **does** support OpenAI organizations with [Zero Data Retention (ZDR)]
OpenAI rejected the request. Error details: Status: 400, Code: unsupported_parameter, Type: invalid_request_error, Message: 400 Previous response cannot be used for this organization due to Zero Data Retention.
```
You may need to upgrade to a more recent version with: `npm i -g @openai/codex@latest`
Ensure you are running `codex` with `--config disable_response_storage=true` or add this line to `~/.codex/config.toml` to avoid specifying the command line option each time:
```toml
disable_response_storage=true
```
See [the configuration documentation on `disable_response_storage`](./codex-rs/config.md#disable_response_storage) for details.
---
@@ -549,51 +533,7 @@ More broadly we welcome contributions - whether you are opening your very first
- Create a _topic branch_ from `main` - e.g. `feat/interactive-prompt`.
- Keep your changes focused. Multiple unrelated fixes should be opened as separate PRs.
-Use `pnpm test:watch` during development for super-fast feedback.
- We use **Vitest** for unit tests, **ESLint** + **Prettier** for style, and **TypeScript** for type-checking.
- Before pushing, run the full test/type/lint suite:
### Git hooks with Husky
This project uses [Husky](https://typicode.github.io/husky/) to enforce code quality checks:
- **Pre-commit hook**: Automatically runs lint-staged to format and lint files before committing
- **Pre-push hook**: Runs tests and type checking before pushing to the remote
These hooks help maintain code quality and prevent pushing code with failing tests. For more details, see [HUSKY.md](./codex-cli/HUSKY.md).
```bash
pnpm test&& pnpm run lint && pnpm run typecheck
```
- If you have **not** yet signed the Contributor License Agreement (CLA), add a PR comment containing the exact text
```text
I have read the CLA Document and I hereby sign the CLA
```
The CLA-Assistant bot will turn the PR status green once all authors have signed.
```bash
# Watch mode (tests rerun on change)
pnpm test:watch
# Type-check without emitting files
pnpm typecheck
# Automatically fix lint + prettier issues
pnpm lint:fix
pnpm format:fix
```
### Debugging
To debug the CLI with a visual debugger, do the following in the `codex-cli` folder:
- Run `pnpm run build` to build the CLI, which will generate `cli.js.map` alongside `cli.js` in the `dist` folder.
- Run the CLI with `node --inspect-brk ./dist/cli.js` The program then waits until a debugger is attached before proceeding. Options:
- In VS Code, choose **Debug: Attach to Node Process** from the command palette and choose the option in the dropdown with debug port `9229` (likely the first option)
- Go to <chrome://inspect> in Chrome and find **localhost:9229** and click **trace**
-Following the [development setup](#development-workflow) instructions above, ensure your change is free of lint warnings and test failures.
### Writing high-impact code changes
@@ -605,7 +545,7 @@ To debug the CLI with a visual debugger, do the following in the `codex-cli` fol
### Opening a pull request
- Fill in the PR template (or include similar information) - **What? Why? How?**
- Run **all** checks locally (`npm test && npm run lint && npm run typecheck`). CI failures that could have been caught locally slow down the process.
- Run **all** checks locally (`cargo test && cargo clippy --tests && cargo fmt -- --config imports_granularity=Item`). CI failures that could have been caught locally slow down the process.
- Make sure your branch is up-to-date with `main` and that you have resolved merge conflicts.
- Mark the PR as **Ready for review** only when you believe it is in a merge-able state.
@@ -652,73 +592,22 @@ The **DCO check** blocks merges until every commit in the PR carries the footer
### Releasing `codex`
To publish a new version of the CLI you first need to stage the npm package. A
helper script in `codex-cli/scripts/` does all the heavy lifting. Inside the
`codex-cli` folder run:
_For admins only._
```bash
# Classic, JS implementation that includes small, native binaries for Linux sandboxing.
pnpm stage-release
Make sure you are on `main` and have no local changes. Then run:
# Optionally specify the temp directory to reuse between runs.
RELEASE_DIR=$(mktemp -d)
pnpm stage-release --tmp "$RELEASE_DIR"
# "Fat" package that additionally bundles the native Rust CLI binaries for
# Linux. End-users can then opt-in at runtime by setting CODEX_RUST=1.
pnpm stage-release --native
```shell
VERSION=0.2.0 # Can also be 0.2.0-alpha.1 or any valid Rust version.
Go to the folder where the release is staged and verify that it works as intended. If so, run the following from the temp folder:
This will make a local commit on top of `main` with `version` set to `$VERSION` in `codex-rs/Cargo.toml` (note that on `main`, we leave the version as `version = "0.0.0"`).
```
cd "$RELEASE_DIR"
npm publish
```
This will push the commit using the tag `rust-v${VERSION}`, which in turn kicks off [the release workflow](.github/workflows/rust-release.yml). This will create a new GitHub Release named `$VERSION`.
### Alternative build options
If everything looks good in the generated GitHub Release, uncheck the **pre-release** box so it is the latest release.
#### Nix flake development
Prerequisite: Nix >= 2.4 with flakes enabled (`experimental-features = nix-command flakes` in `~/.config/nix/nix.conf`).
Enter a Nix development shell:
```bash
# Use either one of the commands according to which implementation you want to work with
nix develop .#codex-cli # For entering codex-cli specific shell
nix develop .#codex-rs # For entering codex-rs specific shell
```
This shell includes Node.js, installs dependencies, builds the CLI, and provides a `codex` command alias.
Build and run the CLI directly:
```bash
# Use either one of the commands according to which implementation you want to work with
nix build .#codex-cli # For building codex-cli
nix build .#codex-rs # For building codex-rs
./result/bin/codex --help
```
Run the CLI via the flake app:
```bash
# Use either one of the commands according to which implementation you want to work with
nix run .#codex-cli # For running codex-cli
nix run .#codex-rs # For running codex-rs
```
Use direnv with flakes
If you have direnv installed, you can use the following `.envrc` to automatically enter the Nix shell when you `cd` into the project directory:
You are a summarization assistant. A conversation follows between a user and a coding-focused AI (Codex). Your task is to generate a clear summary capturing:
• High-level objective or problem being solved
• Key instructions or design decisions given by the user
• Main code actions or behaviors from the AI
• Important variables, functions, modules, or outputs discussed
• Any unresolved questions or next steps
Produce the summary in a structured format like:
**Objective:** …
**User instructions:** … (bulleted)
**AI actions / code behavior:** … (bulleted)
**Important entities:** … (e.g. function names, variables, files)
<p align="center">Lightweight coding agent that runs in your terminal</p>
<p align="center"><code>npm i -g @openai/codex</code></p>
> [!IMPORTANT]
> This is the documentation for the _legacy_ TypeScript implementation of the Codex CLI. It has been superseded by the _Rust_ implementation. See the [README in the root of the Codex repository](https://github.com/openai/codex/blob/main/README.md) for details.

---
<details>
<summary><strong>Table of contents</strong></summary>
Codex CLI is an experimental project under active development. It is not yet stable, may contain bugs, incomplete features, or undergo breaking changes. We're building it in the open with the community and welcome:
- Bug reports
- Feature requests
- Pull requests
- Good vibes
Help us improve by filing issues or submitting PRs (see the section below for how to contribute)!
## Quickstart
Install globally:
```shell
npm install -g @openai/codex
```
Next, set your OpenAI API key as an environment variable:
```shell
exportOPENAI_API_KEY="your-api-key-here"
```
> **Note:** This command sets the key only for your current terminal session. You can add the `export` line to your shell's configuration file (e.g., `~/.zshrc`) but we recommend setting for the session. **Tip:** You can also place your API key into a `.env` file at the root of your project:
>
> ```env
> OPENAI_API_KEY=your-api-key-here
> ```
>
> The CLI will automatically load variables from `.env` (via `dotenv/config`).
<details>
<summary><strong>Use <code>--provider</code> to use other models</strong></summary>
> Codex also allows you to use other providers that support the OpenAI Chat Completions API. You can set the provider in the config file or use the `--provider` flag. The possible options for `--provider` are:
>
> - openai (default)
> - openrouter
> - azure
> - gemini
> - ollama
> - mistral
> - deepseek
> - xai
> - groq
> - arceeai
> - any other provider that is compatible with the OpenAI API
>
> If you use a provider other than OpenAI, you will need to set the API key for the provider in the config file or in the environment variable as:
>
> ```shell
> export <provider>_API_KEY="your-api-key-here"
> ```
>
> If you use a provider not listed above, you must also set the base URL for the provider:
Key flags: `--model/-m`, `--approval-mode/-a`, `--quiet/-q`, and `--notify`.
---
## Memory & project docs
You can give Codex extra instructions and guidance using `AGENTS.md` files. Codex looks for `AGENTS.md` files in the following places, and merges them top-down:
1.`~/.codex/AGENTS.md` - personal global guidance
2.`AGENTS.md` at repo root - shared project notes
3.`AGENTS.md` in the current working directory - sub-folder/feature specifics
Disable loading of these files with `--no-project-doc` or the environment variable `CODEX_DISABLE_PROJECT_DOC=1`.
---
## Non-interactive / CI mode
Run Codex head-less in pipelines. Example GitHub Action step:
```yaml
- name:Update changelog via Codex
run:|
npm install -g @openai/codex
export OPENAI_API_KEY="${{ secrets.OPENAI_KEY }}"
codex -a auto-edit --quiet "update CHANGELOG for next release"
```
Set `CODEX_QUIET_MODE=1` to silence interactive UI noise.
## Tracing / verbose logging
Setting the environment variable `DEBUG=true` prints full API request and response details:
```shell
DEBUG=true codex
```
---
## Recipes
Below are a few bite-size examples you can copy-paste. Replace the text in quotes with your own task. See the [prompting guide](https://github.com/openai/codex/blob/main/codex-cli/examples/prompting_guide.md) for more tips and usage patterns.
<summary>OpenAI released a model called Codex in 2021 - is this related?</summary>
In 2021, OpenAI released Codex, an AI system designed to generate code from natural language prompts. That original Codex model was deprecated as of March 2023 and is separate from the CLI tool.
</details>
<details>
<summary>Which models are supported?</summary>
Any model available with [Responses API](https://platform.openai.com/docs/api-reference/responses). The default is `o4-mini`, but pass `--model gpt-4.1` or set `model: gpt-4.1` in your config file to override.
</details>
<details>
<summary>Why does <code>o3</code> or <code>o4-mini</code> not work for me?</summary>
It's possible that your [API account needs to be verified](https://help.openai.com/en/articles/10910291-api-organization-verification) in order to start streaming responses and seeing chain of thought summaries from the API. If you're still running into issues, please let us know!
</details>
<details>
<summary>How do I stop Codex from editing my files?</summary>
Codex runs model-generated commands in a sandbox. If a proposed command or file change doesn't look right, you can simply type **n** to deny the command or give the model feedback.
</details>
<details>
<summary>Does it work on Windows?</summary>
Not directly. It requires [Windows Subsystem for Linux (WSL2)](https://learn.microsoft.com/en-us/windows/wsl/install) - Codex has been tested on macOS and Linux with Node 22.
</details>
---
## Zero data retention (ZDR) usage
Codex CLI **does** support OpenAI organizations with [Zero Data Retention (ZDR)](https://platform.openai.com/docs/guides/your-data#zero-data-retention) enabled. If your OpenAI organization has Zero Data Retention enabled and you still encounter errors such as:
```
OpenAI rejected the request. Error details: Status: 400, Code: unsupported_parameter, Type: invalid_request_error, Message: 400 Previous response cannot be used for this organization due to Zero Data Retention.
```
You may need to upgrade to a more recent version with: `npm i -g @openai/codex@latest`
---
## Codex open source fund
We're excited to launch a **$1 million initiative** supporting open source projects that use Codex CLI and other OpenAI models.
- Grants are awarded up to **$25,000** API credits.
- Applications are reviewed **on a rolling basis**.
This project is under active development and the code will likely change pretty significantly. We'll update this message once that's complete!
More broadly we welcome contributions - whether you are opening your very first pull request or you're a seasoned maintainer. At the same time we care about reliability and long-term maintainability, so the bar for merging code is intentionally **high**. The guidelines below spell out what "high-quality" means in practice and should make the whole process transparent and friendly.
### Development workflow
- Create a _topic branch_ from `main` - e.g. `feat/interactive-prompt`.
- Keep your changes focused. Multiple unrelated fixes should be opened as separate PRs.
- Use `pnpm test:watch` during development for super-fast feedback.
- We use **Vitest** for unit tests, **ESLint** + **Prettier** for style, and **TypeScript** for type-checking.
- Before pushing, run the full test/type/lint suite:
### Git hooks with Husky
This project uses [Husky](https://typicode.github.io/husky/) to enforce code quality checks:
- **Pre-commit hook**: Automatically runs lint-staged to format and lint files before committing
- **Pre-push hook**: Runs tests and type checking before pushing to the remote
These hooks help maintain code quality and prevent pushing code with failing tests. For more details, see [HUSKY.md](./HUSKY.md).
```bash
pnpm test&& pnpm run lint && pnpm run typecheck
```
- If you have **not** yet signed the Contributor License Agreement (CLA), add a PR comment containing the exact text
```text
I have read the CLA Document and I hereby sign the CLA
```
The CLA-Assistant bot will turn the PR status green once all authors have signed.
```bash
# Watch mode (tests rerun on change)
pnpm test:watch
# Type-check without emitting files
pnpm typecheck
# Automatically fix lint + prettier issues
pnpm lint:fix
pnpm format:fix
```
### Debugging
To debug the CLI with a visual debugger, do the following in the `codex-cli` folder:
- Run `pnpm run build` to build the CLI, which will generate `cli.js.map` alongside `cli.js` in the `dist` folder.
- Run the CLI with `node --inspect-brk ./dist/cli.js` The program then waits until a debugger is attached before proceeding. Options:
- In VS Code, choose **Debug: Attach to Node Process** from the command palette and choose the option in the dropdown with debug port `9229` (likely the first option)
- Go to <chrome://inspect> in Chrome and find **localhost:9229** and click **trace**
### Writing high-impact code changes
1. **Start with an issue.** Open a new one or comment on an existing discussion so we can agree on the solution before code is written.
2. **Add or update tests.** Every new feature or bug-fix should come with test coverage that fails before your change and passes afterwards. 100% coverage is not required, but aim for meaningful assertions.
3. **Document behaviour.** If your change affects user-facing behaviour, update the README, inline help (`codex --help`), or relevant example projects.
4. **Keep commits atomic.** Each commit should compile and the tests should pass. This makes reviews and potential rollbacks easier.
### Opening a pull request
- Fill in the PR template (or include similar information) - **What? Why? How?**
- Run **all** checks locally (`npm test && npm run lint && npm run typecheck`). CI failures that could have been caught locally slow down the process.
- Make sure your branch is up-to-date with `main` and that you have resolved merge conflicts.
- Mark the PR as **Ready for review** only when you believe it is in a merge-able state.
### Review process
1. One maintainer will be assigned as a primary reviewer.
2. We may ask for changes - please do not take this personally. We value the work, we just also value consistency and long-term maintainability.
3. When there is consensus that the PR meets the bar, a maintainer will squash-and-merge.
### Community values
- **Be kind and inclusive.** Treat others with respect; we follow the [Contributor Covenant](https://www.contributor-covenant.org/).
- **Assume good intent.** Written communication is hard - err on the side of generosity.
- **Teach & learn.** If you spot something confusing, open an issue or PR with improvements.
### Getting help
If you run into problems setting up the project, would like feedback on an idea, or just want to say _hi_ - please open a Discussion or jump into the relevant issue. We are happy to help.
Together we can make Codex CLI an incredible tool. **Happy hacking!** :rocket:
### Contributor license agreement (CLA)
All contributors **must** accept the CLA. The process is lightweight:
1. Open your pull request.
2. Paste the following comment (or reply `recheck` if you've signed before):
```text
I have read the CLA Document and I hereby sign the CLA
```
3. The CLA-Assistant bot records your signature in the repo and marks the status check as passed.
No special Git commands, email attachments, or commit footers required.
We provide Codex CLI as a standalone, native executable to ensure a zero-dependency install.
Today, Codex CLI is written in TypeScript and requires Node.js 22+ to run it. For a number of users, this runtime requirement inhibits adoption: they would be better served by a standalone executable. As maintainers, we want Codex to run efficiently in a wide range of environments with minimal overhead. We also want to take advantage of operating system-specific APIs to provide better sandboxing, where possible.
## Installing Codex
To that end, we are moving forward with a Rust implementation of Codex CLI contained in this folder, which has the following benefits:
Today, the easiest way to install Codex is via `npm`, though we plan to publish Codex to other package managers soon.
- The CLI compiles to small, standalone, platform-specific binaries.
- Can make direct, native calls to [seccomp](https://man7.org/linux/man-pages/man2/seccomp.2.html) and [landlock](https://man7.org/linux/man-pages/man7/landlock.7.html) in order to support sandboxing on Linux.
- No runtime garbage collection, resulting in lower memory consumption and better, more predictable performance.
```shell
npm i -g @openai/codex@native
codex
```
Currently, the Rust implementation is materially behind the TypeScript implementation in functionality, so continue to use the TypeScript implementation for the time being. We will publish native executables via GitHub Releases as soon as we feel the Rust version is usable.
You can also download a platform-specific release directly from our [GitHub Releases](https://github.com/openai/codex/releases).
## What's new in the Rust CLI
While we are [working to close the gap between the TypeScript and Rust implementations of Codex CLI](https://github.com/openai/codex/issues/1262), note that the Rust CLI has a number of features that the TypeScript CLI does not!
### Config
Codex supports a rich set of configuration options. Note that the Rust CLI uses `config.toml` instead of `config.json`. See [`config.md`](./config.md) for details.
### Model Context Protocol Support
Codex CLI functions as an MCP client that can connect to MCP servers on startup. See the [`mcp_servers`](./config.md#mcp_servers) section in the configuration documentation for details.
It is still experimental, but you can also launch Codex as an MCP _server_ by running `codex mcp`. Use the [`@modelcontextprotocol/inspector`](https://github.com/modelcontextprotocol/inspector) to try it out:
```shell
npx @modelcontextprotocol/inspector codex mcp
```
### Notifications
You can enable notifications by configuring a script that is run whenever the agent finishes a turn. The [notify documentation](./config.md#notify) includes a detailed example that explains how to get desktop notifications via [terminal-notifier](https://github.com/julienXX/terminal-notifier) on macOS.
### `codex exec` to run Codex programmatially/non-interactively
To run Codex non-interactively, run `codex exec PROMPT` (you can also pass the prompt via `stdin`) and Codex will work on your task until it decides that it is done and exits. Output is printed to the terminal directly. You can set the `RUST_LOG` environment variable to see more about what's going on.
### Use `@` for file search
Typing `@` triggers a fuzzy-filename search over the workspace root. Use up/down to select among the results and Tab or Enter to replace the `@` with the selected path. You can use Esc to cancel the search.
### `--cd`/`-C` flag
Sometimes it is not convenient to `cd` to the directory you want Codex to use as the "working root" before running Codex. Fortunately, `codex` supports a `--cd` option so you can specify whatever folder you want. You can confirm that Codex is honoring `--cd` by double-checking the **workdir** it reports in the TUI at the start of a new session.
### Shell completions
Generate shell completion scripts via:
```shell
codex completion bash
codex completion zsh
codex completion fish
```
### Experimenting with the Codex Sandbox
To test to see what happens when a command is run under the sandbox provided by Codex, we provide the following subcommands in Codex CLI:
```
# macOS
codex debug seatbelt [--full-auto] [COMMAND]...
# Linux
codex debug landlock [--full-auto] [COMMAND]...
```
### Selecting a sandbox policy via `--sandbox`
The Rust CLI exposes a dedicated `--sandbox` (`-s`) flag that lets you pick the sandbox policy **without** having to reach for the generic `-c/--config` option:
```shell
# Run Codex with the default, read-only sandbox
codex --sandbox read-only
# Allow the agent to write within the current workspace while still blocking network access
codex --sandbox workspace-write
# Danger! Disable sandboxing entirely (only do this if you are already running in a container or other isolated env)
codex --sandbox danger-full-access
```
The same setting can be persisted in `~/.codex/config.toml` via the top-level `sandbox_mode = "MODE"` key, e.g. `sandbox_mode = "workspace-write"`.
## Code Organization
@@ -20,330 +94,3 @@ This folder is the root of a Cargo workspace. It contains quite a bit of experim
- [`exec/`](./exec) "headless" CLI for use in automation.
- [`tui/`](./tui) CLI that launches a fullscreen TUI built with [Ratatui](https://ratatui.rs/).
- [`cli/`](./cli) CLI multitool that provides the aforementioned CLIs via subcommands.
## Config
The CLI can be configured via a file named `config.toml`. By default, configuration is read from `~/.codex/config.toml`, though the `CODEX_HOME` environment variable can be used to specify a directory other than `~/.codex`.
The `config.toml` file supports the following options:
### model
The model that Codex should use.
```toml
model="o3"# overrides the default of "codex-mini-latest"
```
### model_provider
Codex comes bundled with a number of "model providers" predefined. This config value is a string that indicates which provider to use. You can also define your own providers via `model_providers`.
For example, if you are running ollama with Mistral locally, then you would need to add the following to your config:
```toml
model="mistral"
model_provider="ollama"
```
because the following definition for `ollama` is included in Codex:
```toml
[model_providers.ollama]
name="Ollama"
base_url="http://localhost:11434/v1"
wire_api="chat"
```
This option defaults to `"openai"` and the corresponding provider is defined as follows:
```toml
[model_providers.openai]
name="OpenAI"
base_url="https://api.openai.com/v1"
env_key="OPENAI_API_KEY"
wire_api="responses"
```
### model_providers
This option lets you override and amend the default set of model providers bundled with Codex. This value is a map where the key is the value to use with `model_provider` to select the correspodning provider.
For example, if you wanted to add a provider that uses the OpenAI 4o model via the chat completions API, then you
```toml
# Recall that in TOML, root keys must be listed before tables.
model="gpt-4o"
model_provider="openai-chat-completions"
[model_providers.openai-chat-completions]
# Name of the provider that will be displayed in the Codex UI.
name="OpenAI using Chat Completions"
# The path `/chat/completions` will be amended to this URL to make the POST
# request for the chat completions.
base_url="https://api.openai.com/v1"
# If `env_key` is set, identifies an environment variable that must be set when
# using Codex with this provider. The value of the environment variable must be
# non-empty and will be used in the `Bearer TOKEN` HTTP header for the POST request.
env_key="OPENAI_API_KEY"
# valid values for wire_api are "chat" and "responses".
wire_api="chat"
```
### approval_policy
Determines when the user should be prompted to approve whether Codex can execute a command:
```toml
# This is analogous to --suggest in the TypeScript Codex CLI
approval_policy="unless-allow-listed"
```
```toml
# If the command fails when run in the sandbox, Codex asks for permission to
# retry the command outside the sandbox.
approval_policy="on-failure"
```
```toml
# User is never prompted: if the command fails, Codex will automatically try
# something out. Note the `exec` subcommand always uses this mode.
approval_policy="never"
```
### profiles
A _profile_ is a collection of configuration values that can be set together. Multiple profiles can be defined in `config.toml` and you can specify the one you
want to use at runtime via the `--profile` flag.
Here is an example of a `config.toml` that defines multiple profiles:
```toml
model="o3"
approval_policy="unless-allow-listed"
sandbox_permissions=["disk-full-read-access"]
disable_response_storage=false
# Setting `profile` is equivalent to specifying `--profile o3` on the command
# line, though the `--profile` flag can still be used to override this value.
profile="o3"
[model_providers.openai-chat-completions]
name="OpenAI using Chat Completions"
base_url="https://api.openai.com/v1"
env_key="OPENAI_API_KEY"
wire_api="chat"
[profiles.o3]
model="o3"
model_provider="openai"
approval_policy="never"
[profiles.gpt3]
model="gpt-3.5-turbo"
model_provider="openai-chat-completions"
[profiles.zdr]
model="o3"
model_provider="openai"
approval_policy="on-failure"
disable_response_storage=true
```
Users can specify config values at multiple levels. Order of precedence is as follows:
2. as part of a profile, where the `--profile` is specified via a CLI (or in the config file itself)
3. as an entry in `config.toml`, e.g., `model = "o3"`
4. the default value that comes with Codex CLI (i.e., Codex CLI defaults to `o4-mini`)
### sandbox_permissions
List of permissions to grant to the sandbox that Codex uses to execute untrusted commands:
```toml
# This is comparable to --full-auto in the TypeScript Codex CLI, though
# specifying `disk-write-platform-global-temp-folder` adds /tmp as a writable
# folder in addition to $TMPDIR.
sandbox_permissions=[
"disk-full-read-access",
"disk-write-platform-user-temp-folder",
"disk-write-platform-global-temp-folder",
"disk-write-cwd",
]
```
To add additional writable folders, use `disk-write-folder`, which takes a parameter (this can be specified multiple times):
```toml
sandbox_permissions=[
# ...
"disk-write-folder=/Users/mbolin/.pyenv/shims",
]
```
### mcp_servers
Defines the list of MCP servers that Codex can consult for tool use. Currently, only servers that are launched by executing a program that communicate over stdio are supported. For servers that use the SSE transport, consider an adapter like [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy).
**Note:** Codex may cache the list of tools and resources from an MCP server so that Codex can include this information in context at startup without spawning all the servers. This is designed to save resources by loading MCP servers lazily.
This config option is comparable to how Claude and Cursor define `mcpServers` in their respective JSON config files, though because Codex uses TOML for its config language, the format is slightly different. For example, the following config in JSON:
```json
{
"mcpServers":{
"server-name":{
"command":"npx",
"args":["-y","mcp-server"],
"env":{
"API_KEY":"value"
}
}
}
}
```
Should be represented as follows in `~/.codex/config.toml`:
```toml
# IMPORTANT: the top-level key is `mcp_servers` rather than `mcpServers`.
[mcp_servers.server-name]
command="npx"
args=["-y","mcp-server"]
env={"API_KEY"="value"}
```
### disable_response_storage
Currently, customers whose accounts are set to use Zero Data Retention (ZDR) must set `disable_response_storage` to `true` so that Codex uses an alternative to the Responses API that works with ZDR:
```toml
disable_response_storage=true
```
### notify
Specify a program that will be executed to get notified about events generated by Codex. Note that the program will receive the notification argument as a string of JSON, e.g.:
```json
{
"type":"agent-turn-complete",
"turn-id":"12345",
"input-messages":["Rename `foo` to `bar` and update the callsites."],
"last-assistant-message":"Rename complete and verified `cargo build` succeeds."
}
```
The `"type"` property will always be set. Currently, `"agent-turn-complete"` is the only notification type that is supported.
As an example, here is a Python script that parses the JSON and decides whether to show a desktop push notification using [terminal-notifier](https://github.com/julienXX/terminal-notifier) on macOS:
print(f"not sending a push notification for: {notification_type}")
return0
subprocess.check_output(
[
"terminal-notifier",
"-title",
title,
"-message",
message,
"-group",
"codex",
"-ignoreDnD",
"-activate",
"com.googlecode.iterm2",
]
)
return0
if__name__=="__main__":
sys.exit(main())
```
To have Codex use this script for notifications, you would configure it via `notify` in `~/.codex/config.toml` using the appropriate path to `notify.py` on your computer:
By default, Codex CLI records messages sent to the model in `$CODEX_HOME/history.jsonl`. Note that on UNIX, the file permissions are set to `o600`, so it should only be readable and writable by the owner.
To disable this behavior, configure `[history]` as follows:
```toml
[history]
persistence="none"# "save-all" is the default value
```
### file_opener
Identifies the editor/URI scheme to use for hyperlinking citations in model output. If set, citations to files in the model output will be hyperlinked using the specified URI scheme so they can be ctrl/cmd-clicked from the terminal to open them.
For example, if the model output includes a reference such as `【F:/home/user/project/main.py†L42-L50】`, then this would be rewritten to link to the URI `vscode://file/home/user/project/main.py:42`.
Note this is **not** a general editor setting (like `$EDITOR`), as it only accepts a fixed set of values:
-`"vscode"` (default)
-`"vscode-insiders"`
-`"windsurf"`
-`"cursor"`
-`"none"` to explicitly disable this feature
Currently, `"vscode"` is the default, though Codex does not verify VS Code is installed. As such, `file_opener` may default to `"none"` or something else in the future.
### project_doc_max_bytes
Maximum number of bytes to read from an `AGENTS.md` file to include in the instructions sent with the first turn of a session. Defaults to 32 KiB.
### tui
Options that are specific to the TUI.
```toml
[tui]
# This will make it so that Codex does not try to process mouse events, which
# means your Terminal's native drag-to-text to text selection and copy/paste
# should work. The tradeoff is that Codex will not receive any mouse events, so
# it will not be possible to use the mouse to scroll conversation history.
#
# Note that most terminals support holding down a modifier key when using the
# mouse to support text selection. For example, even if Codex mouse capture is
# enabled (i.e., this is set to `false`), you can still hold down alt while
To edit files, ALWAYS use the `shell` tool with `apply_patch` CLI. `apply_patch` effectively allows you to execute a diff/patch against a file, but the format of the diff specification is unique to this task, so pay careful attention to these instructions. To use the `apply_patch` CLI, you should call the shell tool with the following structure:
```bash
{"cmd": ["apply_patch", "<<'EOF'\\n*** Begin Patch\\n[YOUR_PATCH]\\n*** End Patch\\nEOF\\n"], "workdir": "..."}
```
Where [YOUR_PATCH] is the actual content of your patch, specified in the following V4A diff format.
*** [ACTION] File: [path/to/file] -> ACTION can be one of Add, Update, or Delete.
For each snippet of code that needs to be changed, repeat the following:
[context_before] -> See below for further instructions on context.
- [old_code] -> Precede the old code with a minus sign.
+ [new_code] -> Precede the new, replacement code with a plus sign.
[context_after] -> See below for further instructions on context.
For instructions on [context_before] and [context_after]:
- By default, show 3 lines of code immediately above and 3 lines immediately below each change. If a change is within 3 lines of a previous change, do NOT duplicate the first change’s [context_after] lines in the second change’s [context_before] lines.
- If 3 lines of context is insufficient to uniquely identify the snippet of code within the file, use the @@ operator to indicate the class or function to which the snippet belongs. For instance, we might have:
@@ class BaseClass
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]
- If a code block is repeated so many times in a class or function such that even a single `@@` statement and 3 lines of context cannot uniquely identify the snippet of code, you can use multiple `@@` statements to jump to the right context. For instance:
@@ class BaseClass
@@ def method():
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]
Note, then, that we do not use line numbers in this diff format, as the context is enough to uniquely identify code. An example of a message that you might pass as "input" to this function, in order to apply a patch, is shown below.
```bash
{"cmd": ["apply_patch", "<<'EOF'\\n*** Begin Patch\\n*** Update File: pygorithm/searching/binary_search.py\\n@@ class BaseClass\\n@@ def search():\\n- pass\\n+ raise NotImplementedError()\\n@@ class Subclass\\n@@ def search():\\n- pass\\n+ raise NotImplementedError()\\n*** End Patch\\nEOF\\n"], "workdir": "..."}
```
File references can only be relative, NEVER ABSOLUTE. After the apply_patch command is run, it will always say "Done!", regardless of whether the patch was successfully applied or not. However, you can determine if there are issue and errors by looking at any warnings or logging lines printed BEFORE the "Done!" is output.
"pr_message":"## Summary\n- add a basic Fibonacci script under `scripts/`\n\n## Testing\n- `node scripts/fibonacci.js 10`\n- `npm run lint` *(fails: next not found)*",
"text":"**Summary**\n\n- Created a command-line Fibonacci script that validates input and prints the result when executed with Node"
},
{
"content_type":"repo_file_citation",
"path":"scripts/fibonacci.js",
"line_range_start":1,
"line_range_end":31
},
{
"content_type":"text",
"text":"\n\n**Testing**\n\n- ❌ `npm run lint` (failed to run `next lint`)"
},
{
"content_type":"terminal_chunk_citation",
"terminal_chunk_id":"7dd543",
"line_range_start":1,
"line_range_end":5
},
{
"content_type":"text",
"text":"\n- ✅ `node scripts/fibonacci.js 10` produced “55”"
},
{
"content_type":"terminal_chunk_citation",
"terminal_chunk_id":"6ee559",
"line_range_start":1,
"line_range_end":3
},
{
"content_type":"text",
"text":"\n\nCodex couldn't run certain commands due to environment limitations. Consider configuring a setup script or internet access in your Codex environment to install dependencies."
Codex supports several mechanisms for setting config values:
- Config-specific command-line flags, such as `--model o3` (highest precedence).
- A generic `-c`/`--config` flag that takes a `key=value` pair, such as `--config model="o3"`.
- The key can contain dots to set a value deeper than the root, e.g. `--config model_providers.openai.wire_api="chat"`.
- Values can contain objects, such as `--config shell_environment_policy.include_only=["PATH", "HOME", "USER"]`.
- For consistency with `config.toml`, values are in TOML format rather than JSON format, so use `{a = 1, b = 2}` rather than `{"a": 1, "b": 2}`.
- If `value` cannot be parsed as a valid TOML value, it is treated as a string value. This means that both `-c model="o3"` and `-c model=o3` are equivalent.
- The `$CODEX_HOME/config.toml` configuration file where the `CODEX_HOME` environment value defaults to `~/.codex`. (Note `CODEX_HOME` will also be where logs and other Codex-related information are stored.)
Both the `--config` flag and the `config.toml` file support the following options:
## model
The model that Codex should use.
```toml
model="o3"# overrides the default of "codex-mini-latest"
```
## model_providers
This option lets you override and amend the default set of model providers bundled with Codex. This value is a map where the key is the value to use with `model_provider` to select the corresponding provider.
For example, if you wanted to add a provider that uses the OpenAI 4o model via the chat completions API, then you could add the following configuration:
```toml
# Recall that in TOML, root keys must be listed before tables.
model="gpt-4o"
model_provider="openai-chat-completions"
[model_providers.openai-chat-completions]
# Name of the provider that will be displayed in the Codex UI.
name="OpenAI using Chat Completions"
# The path `/chat/completions` will be amended to this URL to make the POST
# request for the chat completions.
base_url="https://api.openai.com/v1"
# If `env_key` is set, identifies an environment variable that must be set when
# using Codex with this provider. The value of the environment variable must be
# non-empty and will be used in the `Bearer TOKEN` HTTP header for the POST request.
env_key="OPENAI_API_KEY"
# Valid values for wire_api are "chat" and "responses". Defaults to "chat" if omitted.
wire_api="chat"
# If necessary, extra query params that need to be added to the URL.
# See the Azure example below.
query_params={}
```
Note this makes it possible to use Codex CLI with non-OpenAI models, so long as they use a wire API that is compatible with the OpenAI chat completions API. For example, you could define the following provider to use Codex CLI with Ollama running locally:
```toml
[model_providers.ollama]
name="Ollama"
base_url="http://localhost:11434/v1"
```
Or a third-party provider (using a distinct environment variable for the API key):
```toml
[model_providers.mistral]
name="Mistral"
base_url="https://api.mistral.ai/v1"
env_key="MISTRAL_API_KEY"
```
Note that Azure requires `api-version` to be passed as a query parameter, so be sure to specify it as part of `query_params` when defining the Azure provider:
```toml
[model_providers.azure]
name="Azure"
# Make sure you set the appropriate subdomain for this URL.
env_key="AZURE_OPENAI_API_KEY"# Or "OPENAI_API_KEY", whichever you use.
query_params={api-version="2025-04-01-preview"}
```
It is also possible to configure a provider to include extra HTTP headers with a request. These can be hardcoded values (`http_headers`) or values read from environment variables (`env_http_headers`):
```toml
[model_providers.example]
# name, base_url, ...
# This will add the HTTP header `X-Example-Header` with value `example-value`
# to each request to the model provider.
http_headers={"X-Example-Header"="example-value"}
# This will add the HTTP header `X-Example-Features` with the value of the
# `EXAMPLE_FEATURES` environment variable to each request to the model provider
# _if_ the environment variable is set and its value is non-empty.
The following optional settings control retry behaviour and streaming idle timeouts **per model provider**. They must be specified inside the corresponding `[model_providers.<id>]` block in `config.toml`. (Older releases accepted top‑level keys; those are now ignored.)
Example:
```toml
[model_providers.openai]
name="OpenAI"
base_url="https://api.openai.com/v1"
env_key="OPENAI_API_KEY"
# network tuning overrides (all optional; falls back to built‑in defaults)
request_max_retries=4# retry failed HTTP requests
stream_max_retries=10# retry dropped SSE streams
stream_idle_timeout_ms=300000# 5m idle timeout
```
#### request_max_retries
How many times Codex will retry a failed HTTP request to the model provider. Defaults to `4`.
#### stream_max_retries
Number of times Codex will attempt to reconnect when a streaming response is interrupted. Defaults to `10`.
#### stream_idle_timeout_ms
How long Codex will wait for activity on a streaming response before treating the connection as lost. Defaults to `300_000` (5 minutes).
## model_provider
Identifies which provider to use from the `model_providers` map. Defaults to `"openai"`. You can override the `base_url` for the built-in `openai` provider via the `OPENAI_BASE_URL` environment variable.
Note that if you override `model_provider`, then you likely want to override
`model`, as well. For example, if you are running ollama with Mistral locally,
then you would need to add the following to your config in addition to the new entry in the `model_providers` map:
```toml
model_provider="ollama"
model="mistral"
```
## approval_policy
Determines when the user should be prompted to approve whether Codex can execute a command:
```toml
# Codex has hardcoded logic that defines a set of "trusted" commands.
# Setting the approval_policy to `untrusted` means that Codex will prompt the
# user before running a command not in the "trusted" set.
#
# See https://github.com/openai/codex/issues/1260 for the plan to enable
# end-users to define their own trusted commands.
approval_policy="untrusted"
```
If you want to be notified whenever a command fails, use "on-failure":
```toml
# If the command fails when run in the sandbox, Codex asks for permission to
# retry the command outside the sandbox.
approval_policy="on-failure"
```
If you want the model to run until it decides that it needs to ask you for escalated permissions, use "on-request":
```toml
# The model decides when to escalate
approval_policy="on-request"
```
Alternatively, you can have the model run until it is done, and never ask to run a command with escalated permissions:
```toml
# User is never prompted: if the command fails, Codex will automatically try
# something out. Note the `exec` subcommand always uses this mode.
approval_policy="never"
```
## profiles
A _profile_ is a collection of configuration values that can be set together. Multiple profiles can be defined in `config.toml` and you can specify the one you
want to use at runtime via the `--profile` flag.
Here is an example of a `config.toml` that defines multiple profiles:
```toml
model="o3"
approval_policy="unless-allow-listed"
disable_response_storage=false
# Setting `profile` is equivalent to specifying `--profile o3` on the command
# line, though the `--profile` flag can still be used to override this value.
profile="o3"
[model_providers.openai-chat-completions]
name="OpenAI using Chat Completions"
base_url="https://api.openai.com/v1"
env_key="OPENAI_API_KEY"
wire_api="chat"
[profiles.o3]
model="o3"
model_provider="openai"
approval_policy="never"
model_reasoning_effort="high"
model_reasoning_summary="detailed"
[profiles.gpt3]
model="gpt-3.5-turbo"
model_provider="openai-chat-completions"
[profiles.zdr]
model="o3"
model_provider="openai"
approval_policy="on-failure"
disable_response_storage=true
```
Users can specify config values at multiple levels. Order of precedence is as follows:
2. as part of a profile, where the `--profile` is specified via a CLI (or in the config file itself)
3. as an entry in `config.toml`, e.g., `model = "o3"`
4. the default value that comes with Codex CLI (i.e., Codex CLI defaults to `codex-mini-latest`)
## model_reasoning_effort
If the model name starts with `"o"` (as in `"o3"` or `"o4-mini"`) or `"codex"`, reasoning is enabled by default when using the Responses API. As explained in the [OpenAI Platform documentation](https://platform.openai.com/docs/guides/reasoning?api-mode=responses#get-started-with-reasoning), this can be set to:
-`"low"`
-`"medium"` (default)
-`"high"`
To disable reasoning, set `model_reasoning_effort` to `"none"` in your config:
```toml
model_reasoning_effort="none"# disable reasoning
```
## model_reasoning_summary
If the model name starts with `"o"` (as in `"o3"` or `"o4-mini"`) or `"codex"`, reasoning is enabled by default when using the Responses API. As explained in the [OpenAI Platform documentation](https://platform.openai.com/docs/guides/reasoning?api-mode=responses#reasoning-summaries), this can be set to:
-`"auto"` (default)
-`"concise"`
-`"detailed"`
To disable reasoning summaries, set `model_reasoning_summary` to `"none"` in your config:
By default, `reasoning` is only set on requests to OpenAI models that are known to support them. To force `reasoning` to set on requests to the current model, you can force this behavior by setting the following in `config.toml`:
```toml
model_supports_reasoning_summaries=true
```
## sandbox_mode
Codex executes model-generated shell commands inside an OS-level sandbox.
In most cases you can pick the desired behaviour with a single option:
```toml
# same as `--sandbox read-only`
sandbox_mode="read-only"
```
The default policy is `read-only`, which means commands can read any file on
disk, but attempts to write a file or access the network will be blocked.
A more relaxed policy is `workspace-write`. When specified, the current working directory for the Codex task will be writable (as well as `$TMPDIR` on macOS). Note that the CLI defaults to using the directory where it was spawned as `cwd`, though this can be overridden using `--cwd/-C`.
On macOS (and soon Linux), all writable roots (including `cwd`) that contain a `.git/` folder _as an immediate child_ will configure the `.git/` folder to be read-only while the rest of the Git repository will be writable. This means that commands like `git commit` will fail, by default (as it entails writing to `.git/`), and will require Codex to ask for permission.
```toml
# same as `--sandbox workspace-write`
sandbox_mode="workspace-write"
# Extra settings that only apply when `sandbox = "workspace-write"`.
[sandbox_workspace_write]
# By default, only the cwd for the Codex session will be writable (and $TMPDIR
# on macOS), but you can specify additional writable folders in this array.
writable_roots=["/tmp"]
# Allow the command being run inside the sandbox to make outbound network
# requests. Disabled by default.
network_access=false
```
To disable sandboxing altogether, specify `danger-full-access` like so:
```toml
# same as `--sandbox danger-full-access`
sandbox_mode="danger-full-access"
```
This is reasonable to use if Codex is running in an environment that provides its own sandboxing (such as a Docker container) such that further sandboxing is unnecessary.
Though using this option may also be necessary if you try to use Codex in environments where its native sandboxing mechanisms are unsupported, such as older Linux kernels or on Windows.
## mcp_servers
Defines the list of MCP servers that Codex can consult for tool use. Currently, only servers that are launched by executing a program that communicate over stdio are supported. For servers that use the SSE transport, consider an adapter like [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy).
**Note:** Codex may cache the list of tools and resources from an MCP server so that Codex can include this information in context at startup without spawning all the servers. This is designed to save resources by loading MCP servers lazily.
This config option is comparable to how Claude and Cursor define `mcpServers` in their respective JSON config files, though because Codex uses TOML for its config language, the format is slightly different. For example, the following config in JSON:
```json
{
"mcpServers":{
"server-name":{
"command":"npx",
"args":["-y","mcp-server"],
"env":{
"API_KEY":"value"
}
}
}
}
```
Should be represented as follows in `~/.codex/config.toml`:
```toml
# IMPORTANT: the top-level key is `mcp_servers` rather than `mcpServers`.
[mcp_servers.server-name]
command="npx"
args=["-y","mcp-server"]
env={"API_KEY"="value"}
```
## disable_response_storage
Currently, customers whose accounts are set to use Zero Data Retention (ZDR) must set `disable_response_storage` to `true` so that Codex uses an alternative to the Responses API that works with ZDR:
```toml
disable_response_storage=true
```
## shell_environment_policy
Codex spawns subprocesses (e.g. when executing a `local_shell` tool-call suggested by the assistant). By default it passes **only a minimal core subset** of your environment to those subprocesses to avoid leaking credentials. You can tune this behavior via the **`shell_environment_policy`** block in
`config.toml`:
```toml
[shell_environment_policy]
# inherit can be "core" (default), "all", or "none"
inherit="core"
# set to true to *skip* the filter for `"*KEY*"` and `"*TOKEN*"`
ignore_default_excludes=false
# exclude patterns (case-insensitive globs)
exclude=["AWS_*","AZURE_*"]
# force-set / override values
set={CI="1"}
# if provided, *only* vars matching these patterns are kept
| `inherit` | string | `core` | Starting template for the environment:<br>`core` (`HOME`, `PATH`, `USER`, …), `all` (clone full parent env), or `none` (start empty). |
| `ignore_default_excludes` | boolean | `false` | When `false`, Codex removes any var whose **name** contains `KEY`, `SECRET`, or `TOKEN` (case-insensitive) before other rules run. |
| `exclude` | array<string> | `[]` | Case-insensitive glob patterns to drop after the default filter.<br>Examples: `"AWS_*"`, `"AZURE_*"`. |
| `set` | table<string,string> | `{}` | Explicit key/value overrides or additions – always win over inherited values. |
| `include_only` | array<string> | `[]` | If non-empty, a whitelist of patterns; only variables that match _one_ pattern survive the final step. (Generally used with `inherit = "all"`.) |
The patterns are **glob style**, not full regular expressions: `*` matches any
number of characters, `?` matches exactly one, and character classes like
`[A-Z]`/`[^0-9]` are supported. Matching is always **case-insensitive**. This
syntax is documented in code as `EnvironmentVariablePattern` (see
`core/src/config_types.rs`).
If you just need a clean slate with a few custom entries you can write:
```toml
[shell_environment_policy]
inherit="none"
set={PATH="/usr/bin",MY_FLAG="1"}
```
Currently, `CODEX_SANDBOX_NETWORK_DISABLED=1` is also added to the environment, assuming network is disabled. This is not configurable.
## notify
Specify a program that will be executed to get notified about events generated by Codex. Note that the program will receive the notification argument as a string of JSON, e.g.:
```json
{
"type":"agent-turn-complete",
"turn-id":"12345",
"input-messages":["Rename `foo` to `bar` and update the callsites."],
"last-assistant-message":"Rename complete and verified `cargo build` succeeds."
}
```
The `"type"` property will always be set. Currently, `"agent-turn-complete"` is the only notification type that is supported.
As an example, here is a Python script that parses the JSON and decides whether to show a desktop push notification using [terminal-notifier](https://github.com/julienXX/terminal-notifier) on macOS:
print(f"not sending a push notification for: {notification_type}")
return0
subprocess.check_output(
[
"terminal-notifier",
"-title",
title,
"-message",
message,
"-group",
"codex",
"-ignoreDnD",
"-activate",
"com.googlecode.iterm2",
]
)
return0
if__name__=="__main__":
sys.exit(main())
```
To have Codex use this script for notifications, you would configure it via `notify` in `~/.codex/config.toml` using the appropriate path to `notify.py` on your computer:
By default, Codex CLI records messages sent to the model in `$CODEX_HOME/history.jsonl`. Note that on UNIX, the file permissions are set to `o600`, so it should only be readable and writable by the owner.
To disable this behavior, configure `[history]` as follows:
```toml
[history]
persistence="none"# "save-all" is the default value
```
## file_opener
Identifies the editor/URI scheme to use for hyperlinking citations in model output. If set, citations to files in the model output will be hyperlinked using the specified URI scheme so they can be ctrl/cmd-clicked from the terminal to open them.
For example, if the model output includes a reference such as `【F:/home/user/project/main.py†L42-L50】`, then this would be rewritten to link to the URI `vscode://file/home/user/project/main.py:42`.
Note this is **not** a general editor setting (like `$EDITOR`), as it only accepts a fixed set of values:
-`"vscode"` (default)
-`"vscode-insiders"`
-`"windsurf"`
-`"cursor"`
-`"none"` to explicitly disable this feature
Currently, `"vscode"` is the default, though Codex does not verify VS Code is installed. As such, `file_opener` may default to `"none"` or something else in the future.
## hide_agent_reasoning
Codex intermittently emits "reasoning" events that show the model's internal "thinking" before it produces a final answer. Some users may find these events distracting, especially in CI logs or minimal terminal output.
Setting `hide_agent_reasoning` to `true` suppresses these events in **both** the TUI as well as the headless `exec` sub-command:
```toml
hide_agent_reasoning=true# defaults to false
```
## show_raw_agent_reasoning
Surfaces the model’s raw chain-of-thought ("raw reasoning content") when available.
Notes:
- Only takes effect if the selected model/provider actually emits raw reasoning content. Many models do not. When unsupported, this option has no visible effect.
- Raw reasoning may include intermediate thoughts or sensitive context. Enable only if acceptable for your workflow.
Example:
```toml
show_raw_agent_reasoning=true# defaults to false
```
## model_context_window
The size of the context window for the model, in tokens.
In general, Codex knows the context window for the most common OpenAI models, but if you are using a new model with an old version of the Codex CLI, then you can use `model_context_window` to tell Codex what value to use to determine how much context is left during a conversation.
## model_max_output_tokens
This is analogous to `model_context_window`, but for the maximum number of output tokens for the model.
## project_doc_max_bytes
Maximum number of bytes to read from an `AGENTS.md` file to include in the instructions sent with the first turn of a session. Defaults to 32 KiB.
## tui
Options that are specific to the TUI.
```toml
[tui]
# More to come here
```
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.