## Summary
- add ARC monitor support for MCP tool calls by serializing MCP approval
requests into the ARC action shape and sending the relevant
conversation/policy context to the `/api/codex/safety/arc` endpoint
- route ARC outcomes back into MCP approval flow so `ask-user` falls
back to a user prompt and `steer-model` blocks the tool call, with
guardian/ARC tests covering the new request shape
- update the TUI approval copy from “Approve Once” to “Allow” / “Allow
for this session” and refresh the related
snapshots
---------
Co-authored-by: Fouad Matin <fouad@openai.com>
Co-authored-by: Fouad Matin <169186268+fouad-openai@users.noreply.github.com>
## Summary
- add the OpenAI Docs skill under
codex-rs/skills/src/assets/samples/openai-docs
- include the skill metadata, assets, and GPT-5.4 upgrade reference
files
- exclude the test harness and test fixtures
## Testing
- not run (skill-only asset copy)
This updates the `skill-creator` sample skill to explicitly cover
forward-testing as part of the skill authoring workflow. The guidance
now treats subagent-based validation as a first-class step for complex
or fragile skills, with an emphasis on preserving evaluation integrity
and avoiding leaked context.
The sample initialization script is also updated so newly created skills
point authors toward forward-testing after validation. Together, these
changes make the sample more opinionated about how skills should be
iterated on once the initial implementation is complete.
- Add new guidance to `SKILL.md` on protecting validation integrity,
when to use subagents for forward-testing, and how to structure
realistic test prompts without leaking expected answers.
- Expand the skill creation workflow so iteration explicitly includes
forward-testing for complex skills, including approval guidance for
expensive or risky validation runs.
Fixes a Codex app bug where quitting the app mid-run could leave the
reopened thread stuck in progress and non-interactable. On cold thread
resume, app-server could return an idle thread with a replayed turn
still marked in progress. This marks incomplete replayed turns as
interrupted unless the thread is actually active.
## Summary
- run the split stdout/stderr PTY test through the normal shell helper
on every platform
- use a Windows-native command string instead of depending on Python to
emit split streams
- assert CRLF line endings on Windows explicitly
## Why this fixes the flake
The earlier PTY split-output test used a Python one-liner on Windows
while the rest of the file exercised shell-command behavior. That made
the test depend on runner-local Python availability and masked the real
Windows shell output shape. Using a native cmd-compatible command and
asserting the actual CRLF output makes the split stdout/stderr coverage
deterministic on Windows runners.
Summary
- document output types for the various tool handlers and registry so
the API exposes richer descriptions
- update unified execution helpers and client tests to align with the
new output metadata
- clean up unused helpers across tool dispatch paths
Testing
- Not run (not requested)
There are some bug investigations that currently require us to ask users
for their user ID even though they've already uploaded logs and session
details via `/feedback`. This frustrates users and increases the time
for diagnosis.
This PR includes the ChatGPT user ID in the metadata uploaded for
`/feedback` (both the TUI and app-server).
Replace the Unix shell lookup path in `codex-rs/core/src/shell.rs` to
use
`libc::getpwuid_r()` instead of `libc::getpwuid()` when resolving the
current
user's shell.
Why:
- `getpwuid()` can return pointers into libc-managed shared storage
- on the musl static Linux build, concurrent callers can race on that
storage
- this matches the crash pattern reported in tmux/Linux sessions with
parallel
shell activity
Refs:
- Fixes#13842
Addresses #13586
This doesn't affect our CI scripts. It was user-reported.
Summary
- add `wiremock::ResponseTemplate` and `body_string_contains` imports
behind `#[cfg(not(debug_assertions))]` in
`codex-rs/core/tests/suite/view_image.rs` so release builds only pull
the helpers they actually use
## What changed
- This PR changes only the flaky test setup for
`turn_start_notify_payload_includes_initialize_client_name`.
- Instead of shelling out to `python3` to write the notify payload, the
test uses the first-party `codex-app-server-test-notify-capture` helper.
- The helper writes `notify.json` atomically and the test waits for the
file to exist before reading it.
## Why this fixes the flake
- The old test depended on an external Python interpreter being present
and behaving consistently on every CI runner.
- It also raced the file write: the test could observe the path before
the payload had been fully written, which produced partial reads and
intermittent assertion failures.
- Moving the write into a repo-owned helper removes the external
dependency, and atomic write-plus-wait makes the handoff deterministic.
## Scope
- Test-only change.
## Summary
- stop reserving a localhost port in the websocket tests before spawning
the server
- let the app-server bind `127.0.0.1:0` itself and read back the actual
bound websocket address from stderr
- update the websocket test helpers and callers to use the discovered
address
## Why this fixes the flake
The previous harness reserved a port in the test process, dropped it,
and then asked the server process to bind that same address. On busy
runners there is a race between releasing the reservation and the child
process rebinding it, which can produce sporadic startup failures.
Binding to port `0` inside the server removes that race entirely, and
waiting for the server to report the real bound address makes the tests
connect only after the listener is actually ready.
## Summary
- update the unified exec test to use truncated_output() instead of the
removed output field
- fix the compile failure on latest main after ExecCommandToolOutput
changed shape
## What changed
- The retry test now uses the same streaming SSE test server used by
production-style tests instead of a wiremock sequence.
- The fixture is resolved via `find_resource!`, and the test asserts
that exactly two outbound requests were sent.
## Why this fixes the flake
- The old wiremock sequence approximated early-close behavior, but it
did not reproduce the same streaming semantics the real client sees.
- That meant the retry path depended on mock implementation details
instead of on the actual transport behavior we care about.
- Switching to the streaming SSE helper makes the test exercise the real
early-close/retry contract, and counting requests directly verifies that
we retried exactly once rather than merely hoping the sequence aligned.
## Scope
- Test-only change.
- collect input/output transcript deltas into active handoff transcript
state
- attach and clear that transcript on each handoff, and regenerate
schema/tests
### Purpose
While trying to build out CLI-Tools for the agent to use under skills we
have found that those tools sometimes need to invoke a user elicitation.
These elicitations are handled out of band of the codex app-server but
need to indicate to the exec manager that the command running is not
going to progress on the usual timeout horizon.
### Example
Model calls universal exec:
`$ download-credit-card-history --start-date 2026-01-19 --end-date
2026-02-19 > credit_history.jsonl`
download-cred-card-history might hit a hosted/preauthenticated service
to fetch data. That service might decide that the request requires an
end user approval the access to the personal data. It should be able to
signal to the running thread that the command in question is blocked on
user elicitation. In that case we want the exec to continue, but the
timeout to not expire on the tool call, essentially freezing time until
the user approves or rejects the command at which point the tool would
signal the app-server to decrement the outstanding elicitation count.
Now timeouts would proceed as normal.
### What's Added
- New v2 RPC methods:
- thread/increment_elicitation
- thread/decrement_elicitation
- Protocol updates in:
- codex-rs/app-server-protocol/src/protocol/common.rs
- codex-rs/app-server-protocol/src/protocol/v2.rs
- App-server handlers wired in:
- codex-rs/app-server/src/codex_message_processor.rs
### Behavior
- Counter starts at 0 per thread.
- increment atomically increases the counter.
- decrement atomically decreases the counter; decrement at 0 returns
invalid request.
- Transition rules:
- 0 -> 1: broadcast pause state, pausing all active stopwatches
immediately.
- \>0 -> >0: remain paused.
- 1 -> 0: broadcast unpause state, resuming stopwatches.
- Core thread/session logic:
- codex-rs/core/src/codex_thread.rs
- codex-rs/core/src/codex.rs
- codex-rs/core/src/mcp_connection_manager.rs
### Exec-server stopwatch integration
- Added centralized stopwatch tracking/controller:
- codex-rs/exec-server/src/posix/stopwatch_controller.rs
- Hooked pause/unpause broadcast handling + stopwatch registration:
- codex-rs/exec-server/src/posix/mcp.rs
- codex-rs/exec-server/src/posix/stopwatch.rs
- codex-rs/exec-server/src/posix.rs
## Summary
- move interrupted-turn cleanup so running tasks are aborted before
pending approvals are cleared
- keep unified exec shutdown behavior unchanged
## Why this fixes the flake
The interrupted-turn path could clear pending approvals before the
in-flight task had observed cancellation. On slower runners that let an
approval wait resolve in between those steps, tests would sometimes
surface a model-visible rejection instead of the expected TurnAborted
flow. Draining the active turn first and only then clearing pending
approval state makes the abort ordering deterministic.
Summary
- relocate truncation logic for exec command output into the new
`ExecCommandToolOutput` response helper instead of centralized handler
code
- update all affected tools and unified exec handling to use the new
response item structure and eliminate `Function(FunctionToolOutput)`
responses
- adjust context, registry, and handler interfaces to align with the new
response semantics and error fields
Testing
- Not run (not requested)
**Summary**
- allow `code_mode` to pass enabled tools metadata to the runner and
expose them via `tools.js`
- import tools inside JavaScript rather than relying only on globals or
proxies for nested tool calls
- update specs, docs, and tests to exercise the new bridge and explain
the tooling changes
**Testing**
- Not run (not requested)
(Experimental)
This PR adds a first MVP for hooks, with SessionStart and Stop
The core design is:
- hooks live in a dedicated engine under codex-rs/hooks
- each hook type has its own event-specific file
- hook execution is synchronous and blocks normal turn progression while
running
- matching hooks run in parallel, then their results are aggregated into
a normalized HookRunSummary
On the AppServer side, hooks are exposed as operational metadata rather
than transcript-native items:
- new live notifications: hook/started, hook/completed
- persisted/replayed hook results live on Turn.hookRuns
- we intentionally did not add hook-specific ThreadItem variants
Hooks messages are not persisted, they remain ephemeral. The context
changes they add are (they get appended to the user's prompt)
#### What
###### Context + Problem
With the introduction of plugins, we now have one more type of
`$`-mentionable item in the TUI's popup menu on `$`. Apps, skills, and
plugins can all have the same user-facing name, and we attempt to
distinguish with a category tag suffix, like `[App]`. This has a few
problems:
- We decide to show tags by the text that will be inserted into the
conversation, not the actual user-visible text, so two visibly-identical
entries can have no clarifying category tag suffix
- The category tag is a suffix and commonly gets cut off by long
descriptions
- The skill category tag is currently only displayed on repo skills as
`[Repo]`, which is confusing to most users
- The plugin category tag is currently `[<marketplace-name>]`, which is
also confusing to most users
###### Solution
- **Always** show a **prefix** category tag that is `[Skill]`, `[App]`,
or `[Plugin]`. No conditional rendering or copy.
Before:
<img width="801" height="153" alt="image"
src="https://github.com/user-attachments/assets/448e06e7-2af8-4c14-9804-ed1ca17cf514"
/>
After:
<img width="800" height="118" alt="image"
src="https://github.com/user-attachments/assets/57895b41-06fe-4d92-887b-68704c5a15fd"
/>
I also feel this clarifies the results at-a-glance while you scroll:
https://github.com/user-attachments/assets/cbdd5840-53d9-4656-812c-6e816755e1fd
### Tests
Added + updated tests (including snapshots), tested locally
## Summary
- preserve unknown `:special_path` tokens, including nested entries, so
older Codex builds warn and ignore instead of failing config load
- fail closed with a startup warning when a permissions profile has
missing or empty filesystem entries instead of aborting profile
compilation
- normalize Windows verbatim paths like `\?\C:\...` before absolute-path
validation while keeping explicit errors for truly invalid paths
## Testing
- just fmt
- cargo test -p codex-core permissions_profiles_allow
- cargo test -p codex-core
normalize_absolute_path_for_platform_simplifies_windows_verbatim_paths
- cargo test -p codex-protocol
unknown_special_paths_are_ignored_by_legacy_bridge
- cargo clippy -p codex-core -p codex-protocol --all-targets -- -D
warnings
- cargo clean
## Summary
This is a fast follow to the initial `[permissions]` structure.
- keep the new split-policy carveout behavior for narrower non-write
entries under broader writable roots
- preserve legacy `WorkspaceWrite` semantics by using a cwd-aware bridge
that drops only redundant nested readable roots when projecting from
`SandboxPolicy`
- route the legacy macOS seatbelt adapter through that same legacy
bridge so redundant nested readable roots do not become read-only
carveouts on macOS
- derive the legacy bridge for `command_exec` using the sandbox root cwd
rather than the request cwd so policy derivation matches later sandbox
enforcement
- add regression coverage for the legacy macOS nested-readable-root case
## Examples
### Legacy `workspace-write` on macOS
A legacy `workspace-write` policy can redundantly list a nested readable
root under an already-writable workspace root.
For example, legacy config can effectively mean:
- workspace root (`.` / `cwd`) is writable
- `docs/` is also listed in `readable_roots`
The new shared split-policy helper intentionally treats a narrower
non-write entry under a broader writable root as a carveout for real
`[permissions]` configs. Without this fast follow, the unchanged macOS
seatbelt legacy adapter could project that legacy shape into a
`FileSystemSandboxPolicy` that treated `docs/` like a read-only carveout
under the writable workspace root. In practice, legacy callers on macOS
could unexpectedly lose write access inside `docs/`, even though that
path was writable before the `[permissions]` migration work.
This change fixes that by routing the legacy seatbelt path through the
cwd-aware legacy bridge, so:
- legacy `workspace-write` keeps `docs/` writable when `docs/` was only
a redundant readable root
- explicit `[permissions]` entries like `'.' = 'write'` and `'docs' =
'read'` still make `docs/` read-only, which is the new intended
split-policy behavior
### Legacy `command_exec` with a subdirectory cwd
`command_exec` can run a command from a request cwd that is narrower
than the sandbox root cwd.
For example:
- sandbox root cwd is `/repo`
- request cwd is `/repo/subdir`
- legacy policy is still `workspace-write` rooted at `/repo`
Before this fast follow, `command_exec` derived the legacy bridge using
the request cwd, but the sandbox was later built using the sandbox root
cwd. That mismatch could miss redundant legacy readable roots during
projection and accidentally reintroduce read-only carveouts for paths
that should still be writable under the legacy model.
This change fixes that by deriving the legacy bridge with the same
sandbox root cwd that sandbox enforcement later uses.
## Verification
- `just fmt`
- `cargo test -p codex-core
seatbelt_legacy_workspace_write_nested_readable_root_stays_writable`
- `cargo test -p codex-core test_sandbox_config_parsing`
- `cargo clippy -p codex-core -p codex-app-server --all-targets -- -D
warnings`
- `cargo clean`
## Summary
We need to support allowing request_permissions calls when using
`Reject` policy
<img width="1133" height="588" alt="Screenshot 2026-03-09 at 12 06
40 PM"
src="https://github.com/user-attachments/assets/a8df987f-c225-4866-b8ab-5590960daec5"
/>
Note that this is a backwards-incompatible change for Reject policy. I'm
not sure if we need to add a default based on our current use/setup
## Testing
- [x] Added tests
- [x] Tested locally
## Summary
The apply_patch tool should also respect AdditionalPermissions
## Testing
- [x] Added unit tests
---------
Co-authored-by: Codex <noreply@openai.com>
Healthcheck endpoints for the websocket server
- serve `GET /readyz` and `GET /healthz` from the same listener used for
`--listen ws://...`
- switch the websocket listener over to `axum` upgrade handling instead
of manual socket parsing
- add websocket transport coverage for the health endpoints and document
the new behavior
Testing
- integration tests
- built and tested e2e
```
> curl -i http://127.0.0.1:9234/readyz
HTTP/1.1 200 OK
content-length: 0
date: Fri, 06 Mar 2026 19:20:23 GMT
> curl -i http://127.0.0.1:9234/healthz
HTTP/1.1 200 OK
content-length: 0
date: Fri, 06 Mar 2026 19:20:24 GMT
```
This changes the web_search tool spec in codex-core to use dedicated
Responses-API payload structs instead of shared config types and custom
serializers.
Previously, `ToolSpec::WebSearch` stored `WebSearchFilters` and
`WebSearchUserLocation` directly and relied on hand-written serializers
to shape the outgoing JSON. This worked, but it mixed config/schema
types with the OpenAI Responses payload contract and created an easy
place for drift if those shared types changed later.
### Why
This keeps the boundary clearer:
- app-server/config/schema types stay focused on config
- Responses tool payload types stay focused on the OpenAI wire format
It also makes the serialization behavior obvious from the structs
themselves, instead of hiding it in custom serializer functions.
## Summary
request_permissions flows should support persisting results for the
session.
Open Question: Still deciding if we need within-turn approvals - this
adds complexity but I could see it being useful
## Testing
- [x] Updated unit tests
---------
Co-authored-by: Codex <noreply@openai.com>
## What changed
- TypeScript schema fixture generation now goes through in-memory tree
helpers rather than a heavier on-disk generation path.
- The comparison logic normalizes generated banner and path differences
that are not semantically relevant to the exported schema.
- TypeScript and JSON fixture coverage are split into separate tests,
and the expensive schema-export tests are serialized in `nextest`.
## Why this fixes the flake
- The original fixture coverage mixed several heavy codegen paths into
one monolithic test and then compared generated output that included
incidental banner/path differences.
- On Windows CI, that combination created both runtime pressure and
output variance unrelated to the schema shapes we actually care about.
- Splitting the coverage isolates failures by format, in-memory
generation reduces filesystem churn, normalization strips generator
noise, and serializing the heavy tests removes parallel resource
contention.
## Scope
- Production helper change plus test changes.
## What changed
- The RMCP streamable HTTP tests now wait for metadata and tool
readiness before issuing tool calls.
- OAuth state is isolated per test home.
- The helper server startup path now uses bounded bind retries so
transient `AddrInUse` collisions do not fail the test immediately.
## Why this fixes the flake
- The old tests could begin issuing tool requests before the helper
server had finished advertising its metadata and tools, so the first
request sometimes raced the server startup sequence.
- On top of that, shared OAuth state and occasional bind collisions on
CI runners introduced cross-test environmental noise unrelated to the
functionality under test.
- Readiness polling makes the client wait for an observable “server is
ready” signal, while isolated state and bounded bind retries remove
external contention that was causing intermittent failures.
## Scope
- Test-only change.
This cleans up a bunch of metric plumbing that had started to drift.
The main change is making `codex-otel` the canonical home for shared
metric definitions and metric tag helpers. I moved the `turn/thread`
metric names that were still duplicated into the OTEL metric registry,
added a shared `metrics::tags` module for common tag keys and session
tag construction, and updated `SessionTelemetry` to build its metadata
tags through that shared path.
On the codex-core side, TTFT/TTFM now use the shared metric-name
constants instead of local string definitions. I also switched the
obvious remaining turn/thread metric callsites over to the shared
constants, and added a small helper so TTFT/TTFM can attach an optional
sanitized client.name tag from TurnContext.
This should make follow-on telemetry work less ad hoc:
- one canonical place for metric names
- one canonical place for common metric tag keys/builders
- less duplication between `codex-core` and `codex-otel`
add `plugin/uninstall` app-server endpoint to fully rm plugin from
plugins cache dir and rm entry from user config file.
plugin-enablement is session-scoped, so uninstalls are only picked up in
new sessions (like installs).
added tests.
## Summary
Alternative to #14061 - we need to use a child process on windows to
correctly validate Powershell behavior.
## Testing
- [x] These are tests
## What changed
- add a bounded `resume_until_initial_messages` helper in
`core/tests/suite/resume.rs`
- retry the resume call until `initial_messages` contains the fully
persisted final turn shape before asserting
## Why this fixes flakiness
The old test resumed once immediately after `TurnComplete` and sometimes
read rollout state before the final turn had been persisted. That made
the assertion race persistence timing instead of checking the resumed
message shape. The new helper polls for up to two seconds in 10ms steps
and only asserts once the expected message sequence is actually present,
so the test waits for the real readiness condition instead of depending
on a lucky timing window.
## Scope
- test-only
- no production logic change
## Summary
- align the guardian permission test with the actual sandbox policy it
widens and use a slightly larger Windows-only timeout budget
- expose the additional-permissions normalization helper to the guardian
test module
- replace the guardian popup snapshot assertion with targeted string
assertions
## Why this fixes the flake
This group was carrying two separate sources of drift. The guardian core
test widened derived sandbox policies without updating the source
sandbox policy, and it used a Windows command/timeout combination that
was too tight on slower runners. Separately, the TUI test was
snapshotting the full popup even though unrelated feature text changes
were the only thing moving. The new assertions keep coverage on the
guardian entry itself while removing unrelated snapshot churn.
## Summary
- drain the active turn tasks before clearing pending approvals during
interruption
- keep the turn in hand long enough for interrupted tasks to observe
cancellation first
## Why this fixes the flake
Interrupted turns could clear pending approvals too early, which let an
in-flight approval wait surface as a model-visible rejection before the
turn emitted `TurnAborted`. Reordering the cleanup removes that race
without changing the steady-state task model.
## Summary
- replace the Python-based file creation command in the MCP shell
approval test with native platform commands
- build the expected command string from the exact argv that the test
sends
## Why this fixes the flake
The old test depended on Python startup and shell quoting details that
varied across runners. The new version still verifies the same approval
flow, but it uses `touch` on Unix and `New-Item` on Windows so the
assertion only depends on the MCP shell command that Codex actually
forwards.
Handle cloud requirements 401s with the same auth recovery flow as
normal requests, so permanent refresh failures surface the existing
user-facing auth message instead of a generic workspace-config load
error.
## What changed
- The realtime startup-context tests no longer assume the interesting
websocket payload is always `connection 1 / request 0`.
- Instead, they now wait for the first outbound websocket request that
actually carries `session.instructions`, regardless of which websocket
connection won the accept-order race on the runner.
- The env-key fallback test stays serialized because it mutates process
environment.
## Why this fixes the flake
- The old test synchronized on the mirrored `session.updated` client
event and then inspected a fixed websocket slot.
- On CI, the response websocket and the realtime websocket can race each
other during startup. When the response websocket wins that race, the
fixed slot can contain `response.create` instead of the
startup-context-bearing `session.update` request the test actually cares
about.
- That made the test fail nondeterministically by inspecting the wrong
request, or by timing out waiting on a secondary event even though the
real outbound request path was correct.
- Waiting directly on the first request whose payload includes
`session.instructions` removes both ordering assumptions and makes the
assertion line up with the actual contract under test.
- Separately, serializing the environment-mutating fallback case
prevents unrelated tests from seeing partially updated auth state.
## Scope
- Test-only change.
## What changed
- `snapshot_shell_does_not_inherit_stdin` now runs under its own serial
key.
- The change isolates it from other Unix shell-snapshot tests that also
interact with stdin.
## Why this fixes the flake
- The failure was not a shell-snapshot logic bug. It was shared-stdin
interference between concurrently executing tests.
- When multiple tests compete for inherited stdin at the same time, one
test can observe EOF or consumed input that actually belongs to a
different test.
- Running this specific test in a dedicated serial bucket guarantees
exclusive ownership of stdin, which makes the assertion deterministic
without weakening coverage.
## Scope
- Test-only change.
