## Why
This continues the `codex-tools` migration by moving another passive
tool-spec layer out of `codex-core`.
After `ToolSpec` moved into `codex-tools`, `codex-core` still owned
`ConfiguredToolSpec` and `create_tools_json_for_responses_api()`. Both
are data-model and serialization helpers rather than runtime
orchestration, so keeping them in `core/src/tools/registry.rs` and
`core/src/tools/spec.rs` left passive tool-definition code coupled to
`codex-core` longer than necessary.
## What changed
- moved `ConfiguredToolSpec` into `codex-rs/tools/src/tool_spec.rs`
- moved `create_tools_json_for_responses_api()` into
`codex-rs/tools/src/tool_spec.rs`
- re-exported the new surface from `codex-rs/tools/src/lib.rs`, which
remains exports-only
- updated `core/src/client.rs`, `core/src/tools/registry.rs`, and
`core/src/tools/router.rs` to consume the extracted types and serializer
from `codex-tools`
- moved the tool-list serialization test into
`codex-rs/tools/src/tool_spec_tests.rs`
- added focused unit coverage for `ConfiguredToolSpec::name()`
- simplified `core/src/tools/spec_tests.rs` to use the extracted
`ConfiguredToolSpec::name()` directly and removed the now-redundant
local `tool_name()` helper
- updated `codex-rs/tools/README.md` so the crate boundary reflects the
newly extracted tool-spec wrapper and serialization helper
## Test plan
- `cargo test -p codex-tools`
- `CARGO_TARGET_DIR=/tmp/codex-core-configured-spec cargo test -p
codex-core --lib tools::spec::`
- `CARGO_TARGET_DIR=/tmp/codex-core-configured-spec cargo test -p
codex-core --lib client::`
- `just fix -p codex-tools -p codex-core`
- `just argument-comment-lint`
## References
- #15923
- #15928
- #15944
- #15953
- #16031
- #16047
## Why
macOS BuildBuddy started failing before target analysis because the
Apple CDN object pinned in
[`MODULE.bazel`](fce0f76d57/MODULE.bazel (L28-L36))
now returns `403 Forbidden`. The failure report that triggered this
change was this [BuildBuddy
invocation](https://app.buildbuddy.io/invocation/c57590e0-1bdb-4e19-a86f-74d4a7ded228).
This repo uses `@llvm//extensions:osx.bzl` via `osx.from_archive(...)`,
and that API does not discover a current SDK URL for us. It fetches
exactly the `urls`, `sha256`, and `strip_prefix` we pin. Once Apple
retires that `swcdn.apple.com` object, `@macos_sdk` stops resolving and
every downstream macOS build fails during external repository fetch.
This is the same basic failure mode we hit in
[b9fa08ec61](b9fa08ec61):
the pin itself aged out.
## How I tracked it down
1. I started from the BuildBuddy error and copied the exact
`swcdn.apple.com/.../CLTools_macOSNMOS_SDK.pkg` URL from the failure.
2. I reproduced the issue outside CI by opening that URL directly in a
browser and by running `curl -I` against it locally. Both returned `403
Forbidden`, which ruled out BuildBuddy as the root cause.
3. I searched the repo for that URL and found it hardcoded in
`MODULE.bazel`.
4. I inspected the `llvm` Bzlmod `osx` extension implementation to
confirm that `osx.from_archive(...)` is just a literal fetch of the
pinned archive metadata. There is no automatic fallback or catalog
lookup behind it.
5. I queried Apple's software update catalogs to find the current
Command Line Tools package for macOS 26.x. The useful catalog was:
-
`https://swscan.apple.com/content/catalogs/others/index-26-15-14-13-12-10.16-10.15-10.14-10.13-10.12-10.11-10.10-10.9-mountainlion-lion-snowleopard-leopard.merged-1.sucatalog.gz`
This is scriptable; it does not require opening a website in a browser.
The catalog is a gzip-compressed plist served over HTTP, so the workflow
is just:
1. fetch the catalog,
2. decompress it,
3. search or parse the plist for `CLTools_macOSNMOS_SDK.pkg` entries,
4. inspect the matching product metadata.
The quick shell version I used was:
```shell
curl -L <catalog-url> \
| gzip -dc \
| rg -n -C 6 'CLTools_macOSNMOS_SDK\.pkg|PostDate|English\.dist'
```
That is enough to surface the current product id, package URL, post
date, and the matching `.dist` file. If we want something less
grep-driven next time, the same catalog can be parsed structurally. For
example:
```python
import gzip
import plistlib
import urllib.request
url =
"https://swscan.apple.com/content/catalogs/others/index-26-15-14-13-12-10.16-10.15-10.14-10.13-10.12-10.11-10.10-10.9-mountainlion-lion-snowleopard-leopard.merged-1.sucatalog.gz"
with urllib.request.urlopen(url) as resp:
catalog = plistlib.loads(gzip.decompress(resp.read()))
for product_id, product in catalog["Products"].items():
for package in product.get("Packages", []):
package_url = package.get("URL", "")
if package_url.endswith("CLTools_macOSNMOS_SDK.pkg"):
print(product_id)
print(product.get("PostDate"))
print(package_url)
print(product.get("Distributions", {}).get("English"))
```
In practice, `curl` was only the transport. The important part is that
the catalog itself is a machine-readable plist, so this can be
automated.
6. That catalog contains the newer `047-96692` Command Line Tools
release, and its distribution file identifies it as [Command Line Tools
for Xcode
26.4](https://swdist.apple.com/content/downloads/32/53/047-96692-A_OAHIHT53YB/ybtshxmrcju8m2qvw3w5elr4rajtg1x3y3/047-96692.English.dist).
7. I downloaded that package locally, computed its SHA-256, expanded it
with `pkgutil --expand-full`, and verified that it contains
`Payload/Library/Developer/CommandLineTools/SDKs/MacOSX26.4.sdk`, which
is the correct new `strip_prefix` for this pin.
The core debugging loop looked like this:
```shell
curl -I <stale swcdn URL>
rg 'swcdn\.apple\.com|osx\.from_archive' MODULE.bazel
curl -L <apple 26.x sucatalog> | gzip -dc | rg 'CLTools_macOSNMOS_SDK.pkg'
pkgutil --expand-full CLTools_macOSNMOS_SDK.pkg expanded
find expanded/Payload/Library/Developer/CommandLineTools/SDKs -maxdepth 1 -mindepth 1
```
## What changed
- Updated `MODULE.bazel` to point `osx.from_archive(...)` at the
currently live `047-96692` `CLTools_macOSNMOS_SDK.pkg` object.
- Updated the pinned `sha256` to match that package.
- Updated the `strip_prefix` from `MacOSX26.2.sdk` to `MacOSX26.4.sdk`.
## Verification
- `bazel --output_user_root="$(mktemp -d
/tmp/codex-bazel-sdk-fetch.XXXXXX)" build @macos_sdk//sysroot`
## Notes for next time
As long as we pin raw `swcdn.apple.com` objects, this will likely happen
again. When it does, the expected recovery path is:
1. Reproduce the `403` against the exact URL from CI.
2. Find the stale pin in `MODULE.bazel`.
3. Look up the current CLTools package in the relevant Apple software
update catalog for that macOS major version.
4. Download the replacement package and refresh both `sha256` and
`strip_prefix`.
5. Validate the new pin with a fresh `@macos_sdk` fetch, not just an
incremental Bazel build.
The important detail is that the non-`26` catalog did not surface the
macOS 26.x SDK package here; the `index-26-15-14-...` catalog was the
one that exposed the currently live replacement.
## Why
This continues the `codex-tools` migration by moving another passive
tool-definition layer out of `codex-core`.
After `ResponsesApiTool` and the lower-level schema adapters moved into
`codex-tools`, `core/src/client_common.rs` was still owning `ToolSpec`
and the web-search request wire types even though they are serialized
data models rather than runtime orchestration. Keeping those types in
`codex-core` makes the crate boundary look smaller than it really is and
leaves non-runtime tool-shape code coupled to core.
## What changed
- moved `ToolSpec`, `ResponsesApiWebSearchFilters`, and
`ResponsesApiWebSearchUserLocation` into
`codex-rs/tools/src/tool_spec.rs`
- added focused unit tests in `codex-rs/tools/src/tool_spec_tests.rs`
for:
- `ToolSpec::name()`
- web-search config conversions
- `ToolSpec` serialization for `web_search` and `tool_search`
- kept `codex-rs/tools/src/lib.rs` exports-only by re-exporting the new
module from `lib.rs`
- reduced `core/src/client_common.rs` to a compatibility shim that
re-exports the extracted tool-spec types for current core call sites
- updated `core/src/tools/spec_tests.rs` to consume the extracted
web-search types directly from `codex-tools`
- updated `codex-rs/tools/README.md` so the crate contract reflects that
`codex-tools` now owns the passive tool-spec request models in addition
to the lower-level Responses API structs
## Test plan
- `cargo test -p codex-tools`
- `cargo test -p codex-core --lib tools::spec::`
- `cargo test -p codex-core --lib client_common::`
- `just fix -p codex-tools -p codex-core`
- `just argument-comment-lint`
## References
- #15923
- #15928
- #15944
- #15953
- #16031
## Summary
- remove protocol and core support for discovering and listing custom
prompts
- simplify the TUI slash-command flow and command popup to built-in
commands only
- delete obsolete custom prompt tests, helpers, and docs references
- clean up downstream event handling for the removed protocol events
## Why
`argument-comment-lint` had become a PR bottleneck because the repo-wide
lane was still effectively running a `cargo dylint`-style flow across
the workspace instead of reusing Bazel's Rust dependency graph. That
kept the lint enforced, but it threw away the main benefit of moving
this job under Bazel in the first place: metadata reuse and cacheable
per-target analysis in the same shape as Clippy.
This change moves the repo-wide lint onto a native Bazel Rust aspect so
Linux and macOS can lint `codex-rs` without rebuilding the world
crate-by-crate through the wrapper path.
## What Changed
- add a nightly Rust toolchain with `rustc-dev` for Bazel and a
dedicated crate-universe repo for `tools/argument-comment-lint`
- add `tools/argument-comment-lint/driver.rs` and
`tools/argument-comment-lint/lint_aspect.bzl` so Bazel can run the lint
as a custom `rustc_driver`
- switch repo-wide `just argument-comment-lint` and the Linux/macOS
`rust-ci` lanes to `bazel build --config=argument-comment-lint
//codex-rs/...`
- keep the Python/DotSlash wrappers as the package-scoped fallback path
and as the current Windows CI path
- gate the Dylint entrypoint behind a `bazel_native` feature so the
Bazel-native library avoids the `dylint_*` packaging stack
- update the aspect runtime environment so the driver can locate
`rustc_driver` correctly under remote execution
- keep the dedicated `tools/argument-comment-lint` package tests and
wrapper unit tests in CI so the source and packaged entrypoints remain
covered
## Verification
- `python3 -m unittest discover -s tools/argument-comment-lint -p
'test_*.py'`
- `cargo test` in `tools/argument-comment-lint`
- `bazel build
//tools/argument-comment-lint:argument-comment-lint-driver
--@rules_rust//rust/toolchain/channel=nightly`
- `bazel build --config=argument-comment-lint
//codex-rs/utils/path-utils:all`
- `bazel build --config=argument-comment-lint
//codex-rs/rollout:rollout`
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/16106).
* #16120
* __->__ #16106
This is a follow-up to https://github.com/openai/codex/pull/15922. That
previous PR deleted the old `tui` directory and left the new
`tui_app_server` directory in place. This PR renames `tui_app_server` to
`tui` and fixes up all references.
This PR updates the "PR Babysitter" skill to clarify that non-actionable
review comments should receive a direct reply explaining why no change
is needed, and actionable review comments should be marked "resolved"
after they are addressed.
## Summary
- Moves status surface refresh (`refresh_status_surfaces` /
`refresh_status_line`) from `App` event handlers into `ChatWidget`
setters via a new `refresh_model_dependent_surfaces()` method
- Ensures model-dependent UI stays in sync whenever collaboration mode,
model, or reasoning effort changes, including the footer and terminal
title in both `tui` and `tui_app_server`
- Applies the fix to both `tui` and `tui_app_server` widgets
#15961
## Test plan
- [x] Added snapshot test
`status_line_model_with_reasoning_plan_mode_footer` verifying footer
renders correctly in plan mode
- [x] Added
`terminal_title_model_updates_on_model_change_without_manual_refresh` in
`tui_app_server`
- [ ] Verify switching collaboration modes updates the footer in real
TUI
- [ ] Verify model/reasoning effort changes reflect in the status bar
and terminal title
---------
Co-authored-by: Eric Traut <etraut@openai.com>
## Summary
Add the Windows gnullvm stack-reserve flags to the `*-unit-tests-bin`
path in `codex_rust_crate()`.
## Why
This is the narrow code fix behind the earlier review comment on
[#16067](https://github.com/openai/codex/pull/16067). That comment was
stale relative to the workflow-only PR it landed on, but it pointed at a
real gap in `defs.bzl`.
Today, `codex_rust_crate()` already appends
`WINDOWS_GNULLVM_RUSTC_STACK_FLAGS` for:
- `rust_binary()` targets
- integration-test `rust_test()` targets
But the unit-test binary path still omitted those flags. That meant the
generated `*-unit-tests-bin` executables were not built the same way as
the rest of the Windows gnullvm executables in the macro.
## What Changed
- Added `WINDOWS_GNULLVM_RUSTC_STACK_FLAGS` to the `unit_test_binary`
`rust_test()` rule in `defs.bzl`
- Added a short comment explaining why unit-test binaries need the same
stack-reserve treatment as binaries and integration tests on Windows
gnullvm
## Testing
- `bazel query '//codex-rs/core:*'`
- `bazel query '//codex-rs/shell-command:*'`
Those queries load packages that exercise `codex_rust_crate()`,
including `*-unit-tests-bin` targets. The actual runtime effect is
Windows-specific, so the real end-to-end confirmation still comes from
Windows CI.
## Summary
Split the old all-in-one `rust-ci.yml` into:
- a PR-time Cargo workflow in `rust-ci.yml`
- a full post-merge Cargo workflow in `rust-ci-full.yml`
This keeps the PR path focused on fast Cargo-native hygiene plus the
Bazel `build` / `test` / `clippy` coverage in `bazel.yml`, while moving
the heavyweight Cargo-native matrix to `main`.
## Why
`bazel.yml` is now the main Rust verification workflow for pull
requests. It already covers the Bazel build, test, and clippy signal we
care about pre-merge, and it also runs on pushes to `main` to re-verify
the merged tree and help keep the BuildBuddy caches warm.
What was still missing was a clean split for the Cargo-native checks
that Bazel does not replace yet. The old `rust-ci.yml` mixed together:
- fast hygiene checks such as `cargo fmt --check` and `cargo shear`
- `argument-comment-lint`
- the full Cargo clippy / nextest / release-build matrix
That made every PR pay for the full Cargo matrix even though most of
that coverage is better treated as post-merge verification. The goal of
this change is to leave PRs with the checks we still want before merge,
while moving the heavier Cargo-native matrix off the review path.
## What Changed
- Renamed the old heavyweight workflow to `rust-ci-full.yml` and limited
it to `push` on `main` plus `workflow_dispatch`.
- Added a new PR-only `rust-ci.yml` that runs:
- changed-path detection
- `cargo fmt --check`
- `cargo shear`
- `argument-comment-lint` on Linux, macOS, and Windows
- `tools/argument-comment-lint` package tests when the lint itself or
its workflow wiring changes
- Kept the PR workflow's gatherer as the single required Cargo-native
status so branch protection can stay simple.
- Added `.github/workflows/README.md` to document the intended split
between `bazel.yml`, `rust-ci.yml`, and `rust-ci-full.yml`.
- Preserved the recent Windows `argument-comment-lint` behavior from
`e02fd6e1d3` in `rust-ci-full.yml`, and mirrored cross-platform lint
coverage into the PR workflow.
A few details are deliberate:
- The PR workflow still keeps the Linux lint lane on the
default-targets-only invocation for now, while macOS and Windows use the
broader released-linter path.
- This PR does not change `bazel.yml`; it changes the Cargo-native
workflow around the existing Bazel PR path.
## Testing
- Rebasing this change onto `main` after `e02fd6e1d3`
- `ruby -e 'require "yaml"; %w[.github/workflows/rust-ci.yml
.github/workflows/rust-ci-full.yml .github/workflows/bazel.yml].each {
|f| YAML.load_file(f) }'`
## Why
The initial `argument-comment-lint` rollout left Windows on
default-target coverage because there were still Windows-only callsites
failing under `--all-targets`. This follow-up cleans up those remaining
Windows-specific violations so the Windows CI lane can enforce the same
stricter coverage, leaving Linux as the remaining platform-specific
follow-up.
## What changed
- switched the Windows `rust-ci` argument-comment-lint step back to the
default wrapper invocation so it runs full-target coverage again
- added the required `/*param_name*/` annotations at Windows-gated
literal callsites in:
- `codex-rs/windows-sandbox-rs/src/lib.rs`
- `codex-rs/windows-sandbox-rs/src/elevated_impl.rs`
- `codex-rs/tui_app_server/src/multi_agents.rs`
- `codex-rs/network-proxy/src/proxy.rs`
## Validation
- Windows `argument comment lint` CI on this PR
## Why
We want more of the pre-merge Rust signal to come from `bazel.yml`,
especially on Windows. The Bazel test workflow already exercises
`x86_64-pc-windows-gnullvm`, but the Bazel clippy job still only ran on
Linux x64 and macOS arm64. That left a gap where Windows-only Bazel lint
breakages could slip through until the Cargo-based workflow ran.
This change keeps the fix narrow. Rather than expanding the Bazel clippy
target set or changing the shared setup logic, it extends the existing
clippy matrix to the same Windows GNU toolchain that the Bazel test job
already uses.
## What Changed
- add `windows-latest` / `x86_64-pc-windows-gnullvm` to the `clippy` job
matrix in `.github/workflows/bazel.yml`
- update the nearby workflow comment to explain that the goal is to get
Bazel-native Windows lint coverage on the same toolchain as the Bazel
test lane
- leave the Bazel clippy scope unchanged at `//codex-rs/...
-//codex-rs/v8-poc:all`
## Verification
- parsed `.github/workflows/bazel.yml` successfully with Ruby
`YAML.load_file`
## Why
This PR is the current, consolidated follow-up to the earlier Windows
Bazel attempt in #11229. The goal is no longer just to get a tiny
Windows smoke job limping along: it is to make the ordinary Bazel CI
path usable on `windows-latest` for `x86_64-pc-windows-gnullvm`, with
the same broad `//...` test shape that macOS and Linux already use.
The earlier smoke-list version of this work was useful as a foothold,
but it was not a good long-term landing point. Windows Bazel kept
surfacing real issues outside that allowlist:
- GitHub's Windows runner exposed runfiles-manifest bugs such as
`FINDSTR: Cannot open D:MANIFEST`, which broke Bazel test launchers even
when the manifest file existed.
- `rules_rs`, `rules_rust`, LLVM extraction, and Abseil still needed
`windows-gnullvm`-specific fixes for our hermetic toolchain.
- the V8 path needed more work than just turning the Windows matrix
entry back on: `rusty_v8` does not ship Windows GNU artifacts in the
same shape we need, and Bazel's in-tree V8 build needed a set of Windows
GNU portability fixes.
Windows performance pressure also pushed this toward a full solution
instead of a permanent smoke suite. During this investigation we hit
targets such as `//codex-rs/shell-command:shell-command-unit-tests` that
were much more expensive on Windows because they repeatedly spawn real
PowerShell parsers (see #16057 for one concrete example of that
pressure). That made it much more valuable to get the real Windows Bazel
path working than to keep iterating on a narrowly curated subset.
The net result is that this PR now aims for the same CI contract on
Windows that we already expect elsewhere: keep standalone
`//third_party/v8:all` out of the ordinary Bazel lane, but allow V8
consumers under `//codex-rs/...` to build and test transitively through
`//...`.
## What Changed
### CI and workflow wiring
- re-enable the `windows-latest` / `x86_64-pc-windows-gnullvm` Bazel
matrix entry in `.github/workflows/bazel.yml`
- move the Windows Bazel output root to `D:\b` and enable `git config
--global core.longpaths true` in
`.github/actions/setup-bazel-ci/action.yml`
- keep the ordinary Bazel target set on Windows aligned with macOS and
Linux by running `//...` while excluding only standalone
`//third_party/v8:all` targets from the normal lane
### Toolchain and module support for `windows-gnullvm`
- patch `rules_rs` so `windows-gnullvm` is modeled as a distinct Windows
exec/toolchain platform instead of collapsing into the generic Windows
shape
- patch `rules_rust` build-script environment handling so llvm-mingw
build-script probes do not inherit unsupported `-fstack-protector*`
flags
- patch the LLVM module archive so it extracts cleanly on Windows and
provides the MinGW libraries this toolchain needs
- patch Abseil so its thread-local identity path matches the hermetic
`windows-gnullvm` toolchain instead of taking an incompatible MinGW
pthread path
- keep both MSVC and GNU Windows targets in the generated Cargo metadata
because the current V8 release-asset story still uses MSVC-shaped names
in some places while the Bazel build targets the GNU ABI
### Windows test-launch and binary-behavior fixes
- update `workspace_root_test_launcher.bat.tpl` to read the runfiles
manifest directly instead of shelling out to `findstr`, which was the
source of the `D:MANIFEST` failures on the GitHub Windows runner
- thread a larger Windows GNU stack reserve through `defs.bzl` so
Bazel-built binaries that pull in V8 behave correctly both under normal
builds and under `bazel test`
- remove the no-longer-needed Windows bootstrap sh-toolchain override
from `.bazelrc`
### V8 / `rusty_v8` Windows GNU support
- export and apply the new Windows GNU patch set from
`patches/BUILD.bazel` / `MODULE.bazel`
- patch the V8 module/rules/source layers so the in-tree V8 build can
produce Windows GNU archives under Bazel
- teach `third_party/v8/BUILD.bazel` to build Windows GNU static
archives in-tree instead of aliasing them to the MSVC prebuilts
- reuse the Linux release binding for the experimental Windows GNU path
where `rusty_v8` does not currently publish a Windows GNU binding
artifact
## Testing
- the primary end-to-end validation for this work is the `Bazel`
workflow plus `v8-canary`, since the hard parts are Windows-specific and
depend on real GitHub runner behavior
- before consolidation back onto this PR, the same net change passed the
full Bazel matrix in [run
23675590471](https://github.com/openai/codex/actions/runs/23675590471)
and passed `v8-canary` in [run
23675590453](https://github.com/openai/codex/actions/runs/23675590453)
- those successful runs included the `windows-latest` /
`x86_64-pc-windows-gnullvm` Bazel job with the ordinary `//...` path,
not the earlier Windows smoke allowlist
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/15952).
* #16067
* __->__ #15952
## Why
The `argument-comment-lint` entrypoints had grown into two shell
wrappers with duplicated parsing, environment setup, and Cargo
forwarding logic. The recent `--` separator regression was a good
example of the problem: the behavior was subtle, easy to break, and hard
to verify.
This change rewrites those wrappers in Python so the control flow is
easier to follow, the shared behavior lives in one place, and the tricky
argument/defaulting paths have direct test coverage.
## What changed
- replaced `tools/argument-comment-lint/run.sh` and
`tools/argument-comment-lint/run-prebuilt-linter.sh` with Python
entrypoints: `run.py` and `run-prebuilt-linter.py`
- moved shared wrapper behavior into
`tools/argument-comment-lint/wrapper_common.py`, including:
- splitting lint args from forwarded Cargo args after `--`
- defaulting repo runs to `--manifest-path codex-rs/Cargo.toml
--workspace --no-deps`
- defaulting non-`--fix` runs to `--all-targets` unless the caller
explicitly narrows the target set
- setting repo defaults for `DYLINT_RUSTFLAGS` and `CARGO_INCREMENTAL`
- kept the prebuilt wrapper thin: it still just resolves the packaged
DotSlash entrypoint, keeps `rustup` shims first on `PATH`, infers
`RUSTUP_HOME` when needed, and then launches the packaged `cargo-dylint`
path
- updated `justfile`, `rust-ci.yml`, and
`tools/argument-comment-lint/README.md` to use the Python entrypoints
- updated `rust-ci` so the package job runs Python syntax checks plus
the new wrapper unit tests, and the OS-specific lint jobs invoke the
wrappers through an explicit Python interpreter
This is a follow-up to #16054: it keeps the current lint semantics while
making the wrapper logic maintainable enough to iterate on safely.
## Validation
- `python3 -m py_compile tools/argument-comment-lint/wrapper_common.py
tools/argument-comment-lint/run.py
tools/argument-comment-lint/run-prebuilt-linter.py
tools/argument-comment-lint/test_wrapper_common.py`
- `python3 -m unittest discover -s tools/argument-comment-lint -p
'test_*.py'`
- `python3 ./tools/argument-comment-lint/run-prebuilt-linter.py -p
codex-terminal-detection -- --lib`
- `python3 ./tools/argument-comment-lint/run.py -p
codex-terminal-detection -- --lib`
## Why
`//codex-rs/shell-command:shell-command-unit-tests` became a real
bottleneck in the Windows Bazel lane because repeated calls to
`is_safe_command_windows()` were starting a fresh PowerShell parser
process for every `powershell.exe -Command ...` assertion.
PR #16056 was motivated by that same bottleneck, but its test-only
shortcut was the wrong layer to optimize because it weakened the
end-to-end guarantee that our runtime path really asks PowerShell to
parse the command the way we expect.
This PR attacks the actual cost center instead: it keeps the real
PowerShell parser in the loop, but turns that parser into a long-lived
helper process so both tests and the runtime safe-command path can reuse
it across many requests.
## What Changed
- add `shell-command/src/command_safety/powershell_parser.rs`, which
keeps one mutex-protected parser process per PowerShell executable path
and speaks a simple JSON-over-stdio request/response protocol
- turn `shell-command/src/command_safety/powershell_parser.ps1` into a
long-running parser server with comments explaining the protocol, the
AST-shape restrictions, and why unsupported constructs are rejected
conservatively
- keep request ids and a one-time respawn path so a dead or
desynchronized cached child fails closed instead of silently returning
mixed parser output
- preserve separate parser processes for `powershell.exe` and
`pwsh.exe`, since they do not accept the same language surface
- avoid a direct `PipelineChainAst` type reference in the PowerShell
script so the parser service still runs under Windows PowerShell 5.1 as
well as newer `pwsh`
- make `shell-command/src/command_safety/windows_safe_commands.rs`
delegate to the new parser utility instead of spawning a fresh
PowerShell process for every parse
- add a Windows-only unit test that exercises multiple sequential
requests against the same parser process
## Testing
- adds a Windows-only parser-reuse unit test in `powershell_parser.rs`
- the main end-to-end verification for this change is the Windows CI
lane, because the new service depends on real `powershell.exe` /
`pwsh.exe` behavior
# Summary
Claude Code supports a useful prompt-plus-stdin workflow:
```bash
echo "complex input..." | claude -p "summarize concisely"
```
Codex previously did not support the equivalent `codex exec` form. While
`codex exec` could read the prompt from stdin, it could not combine
piped input with an explicit prompt argument.
This change adds that missing workflow:
```bash
echo "complex input..." | codex exec "summarize concisely"
```
With this change, when `codex exec` receives both a positional prompt
and piped stdin, the prompt remains the instruction and stdin is passed
along as structured `<stdin>...</stdin>` context.
Example:
```bash
curl https://jsonplaceholder.typicode.com/comments \
| ./target/debug/codex exec --skip-git-repo-check "format the top 20 items into a markdown table" \
> table.md
```
This PR also adds regression coverage for:
- prompt argument + piped stdin
- legacy stdin-as-prompt behavior
- `codex exec -` forced-stdin behavior
- empty-stdin error cases
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
`argument-comment-lint` was green in CI even though the repo still had
many uncommented literal arguments. The main gap was target coverage:
the repo wrapper did not force Cargo to inspect test-only call sites, so
examples like the `latest_session_lookup_params(true, ...)` tests in
`codex-rs/tui_app_server/src/lib.rs` never entered the blocking CI path.
This change cleans up the existing backlog, makes the default repo lint
path cover all Cargo targets, and starts rolling that stricter CI
enforcement out on the platform where it is currently validated.
## What changed
- mechanically fixed existing `argument-comment-lint` violations across
the `codex-rs` workspace, including tests, examples, and benches
- updated `tools/argument-comment-lint/run-prebuilt-linter.sh` and
`tools/argument-comment-lint/run.sh` so non-`--fix` runs default to
`--all-targets` unless the caller explicitly narrows the target set
- fixed both wrappers so forwarded cargo arguments after `--` are
preserved with a single separator
- documented the new default behavior in
`tools/argument-comment-lint/README.md`
- updated `rust-ci` so the macOS lint lane keeps the plain wrapper
invocation and therefore enforces `--all-targets`, while Linux and
Windows temporarily pass `-- --lib --bins`
That temporary CI split keeps the stricter all-targets check where it is
already cleaned up, while leaving room to finish the remaining Linux-
and Windows-specific target-gated cleanup before enabling
`--all-targets` on those runners. The Linux and Windows failures on the
intermediate revision were caused by the wrapper forwarding bug, not by
additional lint findings in those lanes.
## Validation
- `bash -n tools/argument-comment-lint/run.sh`
- `bash -n tools/argument-comment-lint/run-prebuilt-linter.sh`
- shell-level wrapper forwarding check for `-- --lib --bins`
- shell-level wrapper forwarding check for `-- --tests`
- `just argument-comment-lint`
- `cargo test` in `tools/argument-comment-lint`
- `cargo test -p codex-terminal-detection`
## Follow-up
- Clean up remaining Linux-only target-gated callsites, then switch the
Linux lint lane back to the plain wrapper invocation.
- Clean up remaining Windows-only target-gated callsites, then switch
the Windows lint lane back to the plain wrapper invocation.
Addresses #15992
The app-server TUI was treating tracked agent threads as closed based on
listener-task bookkeeping that does not reflect live thread state during
normal thread switching. That caused the `/agent` picker to gray out
live agents and could show a false "Agent thread ... is closed" replay
message after switching branches.
This PR fixes the picker refresh path to query the app server for each
tracked thread and derive closed vs loaded state from `thread/read`
status, while preserving cached agent metadata for replay-only threads.
Addresses #16049
`codex resume <name>` and `/resume <name>` could fail in the app-server
TUI path because name lookup pre-filtered `thread/list` with the backend
`search_term`, but saved thread names are hydrated after listing and are
not part of that search index. Resolve names by scanning listed threads
client-side instead, and add a regression test for saved sessions whose
rollout title does not match the thread name.
## Why
Before this change, the SDK CI job built `codex` with Cargo before
running the TypeScript package tests. That step has been getting more
expensive as the Rust workspace grows, while the repo already has a
Bazel-backed build path for the CLI.
The SDK tests also need a normal executable path they can spawn
repeatedly. Moving the job to Bazel exposed an extra CI detail: a plain
`bazel-bin/...` lookup is not reliable under the Linux config because
top-level outputs may stay remote and the wrapper emits status lines
around `cquery` output.
## What Changed
- taught `sdk/typescript/tests/testCodex.ts` to honor `CODEX_EXEC_PATH`
before falling back to the local Cargo-style `target/debug/codex` path
- added `--remote-download-toplevel` to
`.github/scripts/run-bazel-ci.sh` so workflows can force Bazel to
materialize top-level outputs on disk after a build
- switched `.github/workflows/sdk.yml` from `cargo build --bin codex` to
the shared Bazel CI setup and `//codex-rs/cli:codex` build target
- changed the SDK workflow to resolve the built CLI with wrapper-backed
`cquery --output=files`, stage the binary into
`${GITHUB_WORKSPACE}/.tmp/sdk-ci/codex`, and point the SDK tests at that
path via `CODEX_EXEC_PATH`
- kept the warm-up step before Jest and the Bazel repository-cache save
step
## Verification
- `bash -n .github/scripts/run-bazel-ci.sh`
- `./.github/scripts/run-bazel-ci.sh -- cquery --output=files --
//codex-rs/cli:codex | grep -E '^(/|bazel-out/)' | tail -n 1`
- `./.github/scripts/run-bazel-ci.sh --remote-download-toplevel -- build
--build_metadata=TAG_job=sdk -- //codex-rs/cli:codex`
- `CODEX_EXEC_PATH="$PWD/.tmp/sdk-ci/codex" pnpm --dir sdk/typescript
test --runInBand`
- `pnpm --dir sdk/typescript lint`
This is the part 1 of 2 PRs that will delete the `tui` /
`tui_app_server` split. This part simply deletes the existing `tui`
directory and marks the `tui_app_server` feature flag as removed. I left
the `tui_app_server` feature flag in place for now so its presence
doesn't result in an error. It is simply ignored.
Part 2 will rename the `tui_app_server` directory `tui`. I did this as
two parts to reduce visible code churn.
apply_patch sometimes provides additional parent dir as a writable root
when it is already writable. This is mostly a no-op on Mac/Linux but
causes actual ACL churn on Windows that is best avoided. We are also
seeing some actual failures with these ACLs in the wild, which I haven't
fully tracked down, but it's safe/best to avoid doing it altogether.
- [x] Auto / unspecified approval mode: read-only tools now skip before
guardian routing.
- [x] Approve / always-allow mode: read-only tools still skip, now via
the shared early return.
- [x] Prompt mode: read-only tools no longer skip; they continue to
approval.
Addresses #16019
`tui_app_server` renders completed assistant messages from item
notifications, but it only updated `/copy` state from `turn/completed`.
After the app-server migration, turn completion no longer repeats the
final assistant text, so `/copy` could stay unavailable even after the
first normal response.
This PR track the last completed final-answer agent message during an
active app-server turn and promote it into the `/copy` cache when the
turn completes. This restores the pre-migration behavior without
changing rollback handling.
Addresses #15984
HookStarted/HookCompleted notifications were being translated through a
fragile JSON bridge, so hook status/output never reached the renderer.
Early hook notifications could also be dropped during session refresh
before replay.
This PR fixes `tui_app_server` by mapping app-server hook notifications
into TUI hook events explicitly and preserving buffered hook
notifications across refresh, so cold-start and resumed sessions render
the same hook UI as the legacy TUI.
## Why
The previous extraction steps moved shared tool-schema parsing into
`codex-tools`, but `codex-core` still owned the generic Responses API
tool models and the last adapter layer that turned parsed tool
definitions into `ResponsesApiTool` values.
That left `core/src/tools/spec.rs` and `core/src/client_common.rs`
holding a chunk of tool-shaping code that does not need session state,
runtime plumbing, or any other `codex-core`-specific dependency. As a
result, `codex-tools` owned the parsed tool definition, but `codex-core`
still owned the generic wire model that those definitions are converted
into.
This change moves that boundary one step further. `codex-tools` now owns
the reusable Responses/tool wire structs and the shared conversion
helpers for dynamic tools, MCP tools, and deferred MCP aliases.
`codex-core` continues to own `ToolSpec` orchestration and the remaining
web-search-specific request shapes.
## What changed
- added `tools/src/responses_api.rs` to own `ResponsesApiTool`,
`FreeformTool`, `ToolSearchOutputTool`, namespace output types, and the
shared `ToolDefinition -> ResponsesApiTool` adapter helpers
- added `tools/src/responses_api_tests.rs` for deferred-loading
behavior, adapter coverage, and namespace serialization coverage
- rewired `core/src/tools/spec.rs` to use the extracted dynamic/MCP
adapter helpers instead of defining those conversions locally
- rewired `core/src/tools/handlers/tool_search.rs` to use the extracted
deferred MCP adapter and namespace output types directly
- slimmed `core/src/client_common.rs` so it now keeps `ToolSpec` and the
web-search-specific wire types, while reusing the extracted tool models
from `codex-tools`
- moved the extracted seam tests out of `core` and updated
`codex-rs/tools/README.md` plus `tools/src/lib.rs` to reflect the
expanded `codex-tools` boundary
## Test plan
- `cargo test -p codex-tools`
- `cargo test -p codex-core --lib tools::spec::`
- `cargo test -p codex-core --lib tools::handlers::tool_search::`
- `just fix -p codex-tools -p codex-core`
- `just argument-comment-lint`
## References
- [#15923](https://github.com/openai/codex/pull/15923) `codex-tools:
extract shared tool schema parsing`
- [#15928](https://github.com/openai/codex/pull/15928) `codex-tools:
extract MCP schema adapters`
- [#15944](https://github.com/openai/codex/pull/15944) `codex-tools:
extract dynamic tool adapters`
- [#15953](https://github.com/openai/codex/pull/15953) `codex-tools:
introduce named tool definitions`
## Summary
- add `self_serve_business_usage_based` and `enterprise_cbp_usage_based`
to the public/internal plan enums and regenerate the app-server + Python
SDK artifacts
- map both plans through JWT login and backend rate-limit payloads, then
bucket them with the existing Team/Business entitlement behavior in
cloud requirements, usage-limit copy, tooltips, and status display
- keep the earlier display-label remap commit on this branch so the new
Team-like and Business-like plans render consistently in the UI
## Testing
- `just write-app-server-schema`
- `uv run --project sdk/python python
sdk/python/scripts/update_sdk_artifacts.py generate-types`
- `just fix -p codex-protocol -p codex-login -p codex-core -p
codex-backend-client -p codex-cloud-requirements -p codex-tui -p
codex-tui-app-server -p codex-backend-openapi-models`
- `just fmt`
- `just argument-comment-lint`
- `cargo test -p codex-protocol
usage_based_plan_types_use_expected_wire_names`
- `cargo test -p codex-login usage_based`
- `cargo test -p codex-backend-client usage_based`
- `cargo test -p codex-cloud-requirements usage_based`
- `cargo test -p codex-core usage_limit_reached_error_formats_`
- `cargo test -p codex-tui plan_type_display_name_remaps_display_labels`
- `cargo test -p codex-tui remapped`
- `cargo test -p codex-tui-app-server
plan_type_display_name_remaps_display_labels`
- `cargo test -p codex-tui-app-server remapped`
- `cargo test -p codex-tui-app-server
preserves_usage_based_plan_type_wire_name`
## Notes
- a broader multi-crate `cargo test` run still hits unrelated existing
guardian-approval config failures in
`codex-rs/core/src/config/config_tests.rs`
1. Keep curated plugin staging directories under TempDir ownership until
activation succeeds, so failed git/HTTP sync attempts do not leak
plugins-clone-*.
2. Best-effort clean up stale plugins-clone-* directories before
creating a new staged repo, using a conservative age threshold.
3. Emit OTEL counters for curated plugin startup sync transport attempts
and final outcome across git and HTTP paths.
## Why
The TypeScript SDK tests create a fresh `CODEX_HOME` for each Jest case
and delete it during teardown. That cleanup has been flaking because the
real `codex` binary can still be doing background curated-plugin startup
sync under `.tmp/plugins-clone-*`, which races the test harness's
recursive delete and leaves `ENOTEMPTY` failures behind.
This path is unrelated to what the SDK tests are exercising, so letting
plugin startup run during these tests only adds nondeterministic
filesystem activity. This showed up recently in the `sdk` CI lane for
[#16031](https://github.com/openai/codex/pull/16031).
## What Changed
- updated `sdk/typescript/tests/testCodex.ts` to merge test config
through a single helper
- disabled `features.plugins` unconditionally for SDK integration tests
so the CLI does not start curated-plugin sync in the temporary
`CODEX_HOME`
- preserved other explicit feature overrides from individual tests while
forcing `plugins` back to `false`
- kept the existing mock-provider override behavior intact for
SSE-backed tests
## Verification
- `pnpm test --runInBand`
- `pnpm lint`
#15999 introduced a Windows-only `\r\n` mismatch in review-exit template
handling. This PR normalizes those template newlines and separates that
fix from [#16014](https://github.com/openai/codex/pull/16014) so it can
be reviewed independently.
## Why
This continues the `codex-tools` migration by moving one more piece of
generic tool-definition bookkeeping out of `codex-core`.
The earlier extraction steps moved shared schema parsing into
`codex-tools`, but `core/src/tools/spec.rs` still had to supply tool
names separately and perform ad hoc rewrites for deferred MCP aliases.
That meant the crate boundary was still awkward: the parsed shape coming
back from `codex-tools` was missing part of the definition that
`codex-core` ultimately needs to assemble a `ResponsesApiTool`.
This change introduces a named `ToolDefinition` in `codex-tools` so both
MCP tools and dynamic tools cross the crate boundary in the same
reusable model. `codex-core` still owns the final `ResponsesApiTool`
assembly, but less of the generic tool-definition shaping logic stays
behind in `core`.
## What changed
- replaced `ParsedToolDefinition` with a named `ToolDefinition` in
`codex-rs/tools/src/tool_definition.rs`
- added `codex-rs/tools/src/tool_definition_tests.rs` for `renamed()`
and `into_deferred()`
- updated `parse_dynamic_tool()` and `parse_mcp_tool()` to return
`ToolDefinition`
- simplified `codex-rs/core/src/tools/spec.rs` so it adapts
`ToolDefinition` into `ResponsesApiTool` instead of rewriting names and
deferred fields inline
- updated parser tests and `codex-rs/tools/README.md` to reflect the
named tool-definition model
## Test plan
- `cargo test -p codex-tools`
- `cargo test -p codex-core --lib tools::spec::`
## Why
`bazel.yml` already builds and tests the Bazel graph, but `rust-ci.yml`
still runs `cargo clippy` separately. This PR starts the transition to a
Bazel-backed lint lane for `codex-rs` so we can eventually replace the
duplicate Rust build, test, and lint work with Bazel while explicitly
keeping the V8 Bazel path out of scope for now.
To make that lane practical, the workflow also needs to look like the
Bazel job we already trust. That means sharing the common Bazel setup
and invocation logic instead of hand-copying it, and covering the arm64
macOS path in addition to Linux.
Landing the workflow green also required fixing the first lint findings
that Bazel surfaced and adding the matching local entrypoint.
## What changed
- add a reusable `build:clippy` config to `.bazelrc` and export
`codex-rs/clippy.toml` from `codex-rs/BUILD.bazel` so Bazel can run the
repository's existing Clippy policy
- add `just bazel-clippy` so the local developer entrypoint matches the
new CI lane
- extend `.github/workflows/bazel.yml` with a dedicated Bazel clippy job
for `codex-rs`, scoped to `//codex-rs/... -//codex-rs/v8-poc:all`
- run that clippy job on Linux x64 and arm64 macOS
- factor the shared Bazel workflow setup into
`.github/actions/setup-bazel-ci/action.yml` and the shared Bazel
invocation logic into `.github/scripts/run-bazel-ci.sh` so the clippy
and build/test jobs stay aligned
- fix the first Bazel-clippy findings needed to keep the lane green,
including the cross-target `cmsghdr::cmsg_len` normalization in
`codex-rs/shell-escalation/src/unix/socket.rs` and the no-`voice-input`
dead-code warnings in `codex-rs/tui` and `codex-rs/tui_app_server`
## Verification
- `just bazel-clippy`
- `RUNNER_OS=macOS ./.github/scripts/run-bazel-ci.sh -- build
--config=clippy --build_metadata=COMMIT_SHA=local-check
--build_metadata=TAG_job=clippy -- //codex-rs/...
-//codex-rs/v8-poc:all`
- `bazel build --config=clippy
//codex-rs/shell-escalation:shell-escalation`
- `CARGO_TARGET_DIR=/tmp/codex4-shell-escalation-test cargo test -p
codex-shell-escalation`
- `ruby -e 'require "yaml";
YAML.load_file(".github/workflows/bazel.yml");
YAML.load_file(".github/actions/setup-bazel-ci/action.yml")'`
## Notes
- `CARGO_TARGET_DIR=/tmp/codex4-tui-app-server-test cargo test -p
codex-tui-app-server` still hits existing guardian-approvals test and
snapshot failures unrelated to this PR's Bazel-clippy changes.
Related: #15954
## Why
`codex-tools` already owned the shared JSON schema parser and the MCP
tool schema adapter, but `core/src/tools/spec.rs` still parsed dynamic
tools directly.
That left the tool-schema boundary split in two different ways:
- MCP tools flowed through `codex-tools`, while dynamic tools were still
parsed in `codex-core`
- the extracted dynamic-tool path initially introduced a
dynamic-specific parsed shape even though `codex-tools` already had very
similar MCP adapter output
This change finishes that extraction boundary in one step. `codex-core`
still owns `ResponsesApiTool` assembly, but both MCP tools and dynamic
tools now enter that layer through `codex-tools` using the same parsed
tool-definition shape.
## What changed
- added `tools/src/dynamic_tool.rs` and sibling
`tools/src/dynamic_tool_tests.rs`
- introduced `parse_dynamic_tool()` in `codex-tools` and switched
`core/src/tools/spec.rs` to use it for dynamic tools
- added `tools/src/parsed_tool_definition.rs` so both MCP and dynamic
adapters return the same `ParsedToolDefinition`
- updated `core/src/tools/spec.rs` to build `ResponsesApiTool` through a
shared local adapter helper instead of separate MCP and dynamic assembly
paths
- expanded `core/src/tools/spec_tests.rs` so the dynamic-tool adapter
test asserts the full converted `ResponsesApiTool`, including
`defer_loading`
- updated `codex-rs/tools/README.md` to reflect the shared parsed
tool-definition boundary
## Test plan
- `cargo test -p codex-tools`
- `cargo test -p codex-core --lib tools::spec::`
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/15944).
* #15953
* __->__ #15944
## Summary
- split the joined `PATH` before running system `bwrap` lookup
- keep the existing workspace-local `bwrap` skip behavior intact
- add regression tests that exercise real multi-entry search paths
## Why
The PATH-based lookup added in #15791 still wrapped the raw `PATH`
environment value as a single `PathBuf` before passing it through
`join_paths()`. On Unix, a normal multi-entry `PATH` contains `:`, so
that wrapper path is invalid as one path element and the lookup returns
`None`.
That made Codex behave as if no system `bwrap` was installed even when
`bwrap` was available on `PATH`, which is what users in #15340 were
still hitting on `0.117.0-alpha.25`.
## Impact
System `bwrap` discovery now works with normal multi-entry `PATH` values
instead of silently falling back to the vendored binary.
Fixes#15340.
## Validation
- `just fmt`
- `cargo test -p codex-sandboxing`
- `cargo test -p codex-linux-sandbox`
- `just fix -p codex-sandboxing`
- `just argument-comment-lint`
## Why
`codex-utils-pty` and `codex-windows-sandbox` were the remaining crates
in `codex-rs` that still overrode the workspace's Rust 2024 edition.
Moving them forward in a separate PR keeps the baseline edition update
isolated from the follow-on Bazel clippy workflow in #15955, while
making linting and formatting behavior consistent with the rest of the
workspace.
This PR also needs Cargo and Bazel to agree on the edition for
`codex-windows-sandbox`. Without the Bazel-side sync, the experimental
Bazel app-server builds fail once they compile `windows-sandbox-rs`.
## What changed
- switch `codex-rs/utils/pty` and `codex-rs/windows-sandbox-rs` to
`edition = "2024"`
- update `codex-utils-pty` callsites and tests to use the collapsed `if
let` form that Clippy expects under the new edition
- fix the Rust 2024 fallout in `windows-sandbox-rs`, including the
reserved `gen` identifier, `unsafe extern` requirements, and new Clippy
findings that surfaced under the edition bump
- keep the edition bump separate from a larger unsafe cleanup by
temporarily allowing `unsafe_op_in_unsafe_fn` in the Windows entrypoint
modules that now report it under Rust 2024
- update `codex-rs/windows-sandbox-rs/BUILD.bazel` to `crate_edition =
"2024"` so Bazel compiles the crate with the same edition as Cargo
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/15954).
* #15976
* #15955
* __->__ #15954
## Problem
App-server clients could only initiate ChatGPT login through the browser
callback flow, even though the shared login crate already supports
device-code auth. That left VS Code, Codex App, and other app-server
clients without a first-class way to use the existing device-code
backend when browser redirects are brittle or when the client UX wants
to own the login ceremony.
## Mental model
This change adds a second ChatGPT login start path to app-server:
clients can now call `account/login/start` with `type:
"chatgptDeviceCode"`. App-server immediately returns a `loginId` plus
the device-code UX payload (`verificationUrl` and `userCode`), then
completes the login asynchronously in the background using the existing
`codex_login` polling flow. Successful device-code login still resolves
to ordinary `chatgpt` auth, and completion continues to flow through the
existing `account/login/completed` and `account/updated` notifications.
## Non-goals
This does not introduce a new auth mode, a new account shape, or a
device-code eligibility discovery API. It also does not add automatic
fallback to browser login in core; clients remain responsible for
choosing when to request device code and whether to retry with a
different UX if the backend/admin policy rejects it.
## Tradeoffs
We intentionally keep `login_chatgpt_common` as a local validation
helper instead of turning it into a capability probe. Device-code
eligibility is checked by actually calling `request_device_code`, which
means policy-disabled cases surface as an immediate request error rather
than an async completion event. We also keep the active-login state
machine minimal: browser and device-code logins share the same public
cancel contract, but device-code cancellation is implemented with a
local cancel token rather than a larger cross-crate refactor.
## Architecture
The protocol grows a new `chatgptDeviceCode` request/response variant in
app-server v2. On the server side, the new handler reuses the existing
ChatGPT login precondition checks, calls `request_device_code`, returns
the device-code payload, and then spawns a background task that waits on
either cancellation or `complete_device_code_login`. On success, it
reuses the existing auth reload and cloud-requirements refresh path
before emitting `account/login/completed` success and `account/updated`.
On failure or cancellation, it emits only `account/login/completed`
failure. The existing `account/login/cancel { loginId }` contract
remains unchanged and now works for both browser and device-code
attempts.
## Tests
Added protocol serialization coverage for the new request/response
variant, plus app-server tests for device-code success, failure, cancel,
and start-time rejection behavior. Existing browser ChatGPT login
coverage remains in place to show that the callback-based flow is
unchanged.
