Give saturated remote full-ci enough time for turn startup, tool output polling, and rollback events to reach their existing assertions.
Co-authored-by: Codex <noreply@openai.com>
Honor pre-cancelled delegate spawns before starting work, wait for TurnStarted in the remote no-wait helper, and give serialized rollback integration tests matching nextest headroom.
Co-authored-by: Codex <noreply@openai.com>
Keep remote routing tests from stalling before the first model request by consuming Codex events while waiting for the captured tool output.
Co-authored-by: Codex <noreply@openai.com>
Allow the serialized Docker-backed remote environment tests enough test-level time to reach their existing tool-output assertions under CI load.
Co-authored-by: Codex <noreply@openai.com>
Keep the assertions intact while allowing rollback and remote tool-output events longer to arrive under the remote full-ci load.
Co-authored-by: Codex <noreply@openai.com>
Add an optional captured-response helper for function call output values so remote-routing tests can poll without panicking on earlier requests.
Co-authored-by: Codex <noreply@openai.com>
function_call_output already returns an owned JSON value, so the remote-routing wait helper should return it directly instead of calling cloned().
Co-authored-by: Codex <noreply@openai.com>
The remote full-ci failures are now concentrated in tests that wait for unrelated completion signals under saturated CI. The remote routing tests only assert the tool output, so submit the turn without waiting for final turn completion and poll the captured function_call_output directly. Rollback-heavy tests are serialized and use an explicit event timeout for their rollback notification.
Co-authored-by: Codex <noreply@openai.com>
Remote full-ci runs these Docker-backed environment tests against one shared exec-server, so serialize the remote-env cases and make the rollback/event assertions tolerate CI startup latency. The unified-exec network-denial tests now still require a failed end event but also accept the remote exec-server startup failure observed when the minimal container cannot start the probe command.
Co-authored-by: Codex <noreply@openai.com>
The remote full-ci stabilization changed TestCodex turn-start waiting to use wait_for_event_with_timeout directly, leaving the wait_for_event_match import unused under -D warnings.
Co-authored-by: Codex <noreply@openai.com>
Remote full-ci was past the project-root failures but still exposed remote-test startup assumptions. Emit failed unified-exec end events when process creation is rejected, make the network-denial tests use /bin/sh in minimal remote containers, align TurnStarted waits with the existing submit-turn completion timeout, and make the delayed realtime sideband handshake deterministic.
Co-authored-by: Codex <noreply@openai.com>
Remote test workspaces can live only inside the executor filesystem, so local git discovery cannot find repo markers for turn diffs. Add an executor-filesystem-backed repo root lookup and use it when building the turn diff display root. Align core-skills test project-layer discovery with the same ambient git marker filtering used by production discovery.
Co-authored-by: Codex <noreply@openai.com>
Remote CI containers can expose an ambient /.git marker. Treat filesystem-root .git markers like stale /tmp markers so test tempdirs do not inherit host/container repo state, while preserving nested real repos.
Co-authored-by: Codex <noreply@openai.com>
World-writable sticky directories such as /tmp should not become implicit project roots just because a stale .git marker exists there. Continue walking past those markers while still honoring real repositories below them.
Co-authored-by: Codex <noreply@openai.com>
## Summary
- make EnvironmentProvider::snapshot path-free and keep providers
focused on provider-owned remote environments
- let provider snapshots request local inclusion via include_local, with
environments.toml including local and CODEX_EXEC_SERVER_URL excluding
local
- move reserved local environment construction into EnvironmentManager
using ExecServerRuntimePaths
Follow-up to https://github.com/openai/codex/pull/20667
## Testing
- just fmt
- git diff --check
- devbox: bazel build --bes_backend= --bes_results_url=
//codex-rs/exec-server:exec-server
- devbox: bazel test --bes_backend= --bes_results_url=
//codex-rs/exec-server:exec-server-unit-tests
Co-authored-by: Codex <noreply@openai.com>
## Summary
Startup tool construction currently depends on connector directory
metadata for `tool_suggest` discoverables. On a cold directory cache,
that can put slow connector-directory requests on the blocking path even
though the tools array only needs directory data for install
suggestions, not for the live connector MCP tools themselves.
This PR keeps the discoverables path off that cold network fetch:
- read connector directory metadata from cache only when building
discoverable tools
- persist connector directory metadata to
`~/.codex/cache/codex_app_directory/<hash>.json` and use it to hydrate
the in-memory cache on later runs before the normal refresh path updates
it
- use connector-directory-specific cache naming to distinguish this
metadata cache from the separate Codex Apps tools-spec cache
This reduces first-turn startup work without changing how live connector
MCP tools are sourced. Longer term, directory-backed install suggestions
should move to a search-based flow so they no longer need to be inlined
into the tools prompt at all.
## Testing
- `cargo test -p codex-connectors`
- `cargo test -p codex-chatgpt`
- `cargo test -p codex-core
request_plugin_install_is_available_without_search_tool_after_discovery_attempts`
- `cargo test -p codex-core
tool_suggest_uses_connector_id_fallback_when_directory_cache_is_empty`
## Summary
In https://github.com/openai/codex/pull/21584, we disabled doctests for
crates that lack any doctests. We can enforce that property via `cargo
shear --deny-warnings`: crates that lack doctests will be flagged if
doctests are enabled, and crates with doctests will be flagged if
doctests are disabled.
A few additional notes:
- By adding `--deny-warnings`, `cargo shear` also flagged a number of
modules that were not reachable at all. Some of those have been removed.
- This PR removes a usage of `windows_modules!` (since `cargo shear` and
`rustfmt` couldn't see through it) in favor of simple `#[cfg(target_os =
"windows")]` macros. As a consequence, many of these files exhibit churn
in this PR, since they weren't being formatted by `rustfmt` at all on
main.
- Again, to make the code more analyzable, this PR also removes some
usages of `#[path = "cwd_junction.rs"]` in favor of a more standard
module structure. The bin sidecar structure is still retained, but,
e.g., `windows-sandbox-rs/src/bin/command_runner.rs` was moved to
`windows-sandbox-rs/src/bin/command_runner/main.rs`, and so on.
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
The legacy `AfterToolUse` hook path was still wired through core tool
dispatch even though the hooks registry never populated any handlers for
it. The supported hook surface is `PostToolUse`, so the old
infrastructure was dead code on the hot path.
## What changed
- Removed the legacy `AfterToolUse` dispatch from `codex-core` tool
execution.
- Removed the unused legacy hook payload types and exports from
`codex-hooks`.
- Simplified legacy notify handling now that `HookEvent` only carries
`AfterAgent`.
## Validation
- `cargo test -p codex-hooks`
- `cargo test -p codex-core registry`
## Why
`apply_patch` is now a freeform/custom tool. Keeping the old
JSON/function-style registration and parsing path left another way for
models and tests to invoke `apply_patch`, which made the tool surface
harder to reason about.
## What changed
- Removed the `ApplyPatchToolType::Function` variant, JSON `apply_patch`
spec, and handler support for function payloads.
- Kept `apply_patch_tool_type = freeform` as the supported model
metadata path, including Bedrock catalog metadata.
- Migrated `apply_patch` tests and SSE fixtures to custom/freeform tool
calls.
## Verification
- `cargo test -p codex-tools -p codex-protocol -p codex-model-provider`
- `cargo test -p codex-core tools::handlers::apply_patch --lib`
- `cargo test -p codex-core --test all
apply_patch_tool_executes_and_emits_patch_events`
- `cargo test -p codex-core --test all
apply_patch_reports_parse_diagnostics`
- `cargo test -p codex-exec test_apply_patch_tool`
- `just fix -p codex-core`
- `just fix -p codex-tools -p codex-protocol -p codex-model-provider -p
codex-exec`
## Summary
TL;DR: teaches `codex-rs` / app-server to request a desktop-provided
attestation token and attach it as `x-oai-attestation` on the scoped
ChatGPT Codex request paths.
## Details
This PR teaches the Codex app-server runtime how to request and attach
an attestation token. It does not generate DeviceCheck tokens directly;
instead, it relies on the connected desktop app to advertise that it can
generate attestation and then asks that app for a fresh header value
when needed.
The flow is:
1. The Codex desktop app connects to app-server.
2. During `initialize`, the app can advertise that it supports
`requestAttestation`.
3. Before app-server calls selected ChatGPT Codex endpoints, it sends
the internal server request `attestation/generate` to the app.
4. app-server receives a pre-encoded header value back.
5. app-server forwards that value as `x-oai-attestation` on the scoped
outbound requests.
The code in this repo is mostly protocol and runtime plumbing: it adds
the app-server request/response shape, introduces an attestation
provider in core, wires that provider into Responses / compaction /
realtime setup paths, and covers the intended scoping with tests. The
signed macOS DeviceCheck generation remains owned by the desktop app PR.
## Related PR
- Codex desktop app implementation:
https://github.com/openai/openai/pull/878649
## Validation
<details>
<summary>Tests run</summary>
```sh
cargo test -p codex-app-server-protocol
cargo test -p codex-core attestation --lib
cargo test -p codex-app-server --lib attestation
```
Also ran:
```sh
just fix -p codex-core
just fix -p codex-app-server
just fix -p codex-app-server-protocol
just fmt
just write-app-server-schema
```
</details>
<details>
<summary>E2E DeviceCheck validation</summary>
First validated the signed desktop app boundary directly: launched a
packaged signed `Codex.app`, sent `attestation/generate`, decoded the
returned `v1.` attestation header, and validated the extracted
DeviceCheck token with `personal/jm/verify_devicecheck_token.py` using
bundle ID `com.openai.codex`. Apple returned `status_code: 200` and
`is_ok: true`.
Then ran the fuller app + app-server flow. The packaged `Codex.app`
launched a current-branch app-server via `CODEX_CLI_PATH`, and a local
MITM proxy intercepted outbound `chatgpt.com` traffic. The app-server
requested `attestation/generate` from the real Electron app process, and
the intercepted `/backend-api/codex/responses` traffic included
`x-oai-attestation` on both routes:
```text
GET /backend-api/codex/responses Upgrade: websocket x-oai-attestation: present
POST /backend-api/codex/responses Upgrade: none x-oai-attestation: present
```
The captured header decoded to a DeviceCheck token that also validated
with Apple for `com.openai.codex` (`status_code: 200`, `is_ok: true`,
team `2DC432GLL2`).
</details>
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
`ToolName::display()` made it too easy to flatten tool identity and
accidentally compare rendered strings. Tool identity should stay
structural until a legacy string boundary actually requires the
flattened spelling.
## What
- Removes `ToolName::display()` and relies on the existing `Display`
impl for messages and errors.
- Adds structural ordering for `ToolName` and uses it for
sorting/deduping deferred tools.
- Carries `ToolName` through tool/sandbox plumbing, flattening only at
legacy boundaries such as hook payloads, telemetry tags, and Responses
tool names.
- Updates MCP normalization tests to assert `ToolName` structure instead
of rendered strings.
## Testing
- `cargo test -p codex-mcp test_normalize_tools`
- `cargo test -p codex-core unavailable_tool`
- `just fix -p codex-protocol`
- `just fix -p codex-mcp`
- `just fix -p codex-core`
## Why
Codex assisted-code attribution needs a client-side accepted-code source
that does not upload raw code. This adds a hash-only analytics event
derived from the turn diff so downstream attribution can compare
accepted Codex lines against commit or PR diffs.
## What Changed
- Parse accepted/effective added lines from the final turn diff and emit
`codex_accepted_line_fingerprints` analytics.
- Hash repo, path, and normalized line content before upload; raw code
and raw diffs are not included in the event.
- Chunk large fingerprint payloads and send accepted-line fingerprint
events in isolated requests while preserving normal batching for other
analytics events.
- Canonicalize Git remote URLs before repo hashing so SSH/HTTPS GitHub
remotes join to the same repo hash.
- Add parser coverage for unified diff hunk lines that look like `+++`
or `---` file headers.
## Verification
- `cargo test -p codex-analytics`
- `cargo test -p codex-git-utils canonicalize_git_remote_url`
- `just fix -p codex-analytics`
- `just bazel-lock-check`
- `git diff --check`
## Why
The earlier PRs add stdio transport support and the config-backed
environment provider, but the feature remains inert until normal Codex
entrypoints construct `EnvironmentManager` with enough context to
discover `CODEX_HOME/environments.toml`. This final stack PR activates
the provider while preserving the legacy `CODEX_EXEC_SERVER_URL`
fallback when no environments file exists.
**Stack position:** this is PR 5 of 5. It is the product wiring PR that
activates the configured environment provider added in PR 4.
## What Changed
- Thread `codex_home` into `EnvironmentManagerArgs`.
- Change `EnvironmentManager::new(...)` to load the provider from
`CODEX_HOME`.
- Preserve legacy behavior by falling back to
`DefaultEnvironmentProvider::from_env()` when `environments.toml` is
absent.
- Make `environments.toml`-backed managers start new threads with all
configured environments, default first, while keeping the legacy env-var
path single-default.
- Update the app-server, TUI, exec, MCP server, connector, prompt-debug,
and thread-manager-sample callsites to pass `codex_home` and handle
provider-loading errors.
## Self-Review Notes
- The multi-environment startup path is intentionally tied to the
`environments.toml` provider. Using `>1` configured environment as the
only signal would also expand the legacy `CODEX_EXEC_SERVER_URL`
provider because it keeps `local` addressable alongside `remote`.
- The startup environment list is still derived inside
`EnvironmentManager`; the provider only says whether its snapshot should
start new threads with all configured environments.
- The thread-manager sample was updated to pass the current
`ThreadManager::new(...)` installation id argument so the stack compiles
under Bazel.
## Stack
- 1. https://github.com/openai/codex/pull/20663 - Add stdio exec-server
listener
- 2. https://github.com/openai/codex/pull/20664 - Add stdio exec-server
client transport
- 3. https://github.com/openai/codex/pull/20665 - Make environment
providers own default selection
- 4. https://github.com/openai/codex/pull/20666 - Add CODEX_HOME
environments TOML provider
- **5. This PR:** https://github.com/openai/codex/pull/20667 - Load
configured environments from CODEX_HOME
Split from original draft: https://github.com/openai/codex/pull/20508
## Validation
- `just fmt`
- `git diff --check`
- `bazel build --config=remote --strategy=remote
--remote_download_toplevel
//codex-rs/thread-manager-sample:codex-thread-manager-sample`
- `bazel test --config=remote --strategy=remote
--remote_download_toplevel
//codex-rs/exec-server:exec-server-unit-tests`
- `bazel test --config=remote --strategy=remote
--remote_download_toplevel --test_sharding_strategy=disabled
--test_arg=default_thread_environment_selections_use_manager_default_id
//codex-rs/core:core-unit-tests`
- `bazel test --config=remote --strategy=remote
--remote_download_toplevel --test_sharding_strategy=disabled
--test_arg=start_thread_uses_all_default_environments_from_codex_home
//codex-rs/core:core-unit-tests`
## Documentation
This activates `CODEX_HOME/environments.toml`; user-facing documentation
should be added before this stack is treated as a documented public
workflow.
---------
Co-authored-by: Codex <noreply@openai.com>
* Pass installation ID for storage on enrollments server for
deduping/grouping multiple appservers per installation
* Pass installation ID in remoteControl/status/changed events
## Why
`/fast` was wired as a one-off slash command even though model metadata
now exposes service tiers as catalog data. That meant adding another
tier, such as a slower/cheaper tier, would require more hardcoded TUI
plumbing instead of letting the model catalog drive the available
commands.
This change makes service-tier commands data-driven: each advertised
`service_tiers` entry becomes a `/name` command using the catalog
description, while the request path sends the tier `id` only when the
selected model supports it.
## What Changed
- Removed the hardcoded `/fast` slash-command variant and introduced
dynamic service-tier command items in the composer and command popup.
- Added toggle behavior for service-tier commands: invoking `/name`
selects that tier, and invoking it again clears the selection.
- Preserved the existing Fast-mode keybinding/status affordances by
resolving the current model tier whose name is `fast`, while still
sending the tier request value such as `priority`.
- Persisted service-tier selections as raw request strings so non-fast
tiers can round-trip through config.
- Updated the Bedrock catalog entry to advertise fast support through
`service_tiers` with `id: "priority"` and `name: "fast"`.
- Added defensive filtering in core so unsupported selected service
tiers are omitted from `/responses` requests.
## Validation
- Added/updated coverage for dynamic service-tier slash command lookup,
popup descriptions, composer dispatch, TUI fast toggling, and
unsupported-tier omission in core request construction.
- Local tests were not run per request.
---------
Co-authored-by: Codex <noreply@openai.com>
Since https://github.com/openai/codex/pull/21255, `rust-ci-full` has
been failing due to a missing `bwrap`.
```
thread 'main' panicked at linux-sandbox/src/launcher.rs:43:13:
bubblewrap is unavailable: no system bwrap was found on PATH and no bundled codex-resources/bwrap binary was found next to the Codex executable
```
Since the happy path is now to use the system binary, let's ensure
that's installed.
8d51826631
was necessary for the `bwrap` executable to be discoverable when the
working directory is `/`.
I ran `rust-ci-full` at
https://github.com/openai/codex/actions/runs/25528074506
---------
Co-authored-by: Codex <noreply@openai.com>
## Summary
1. Removes the broad `DARWIN_USER_CACHE_DIR` write rule from the macOS
Seatbelt network policy.
2. Removes the now unused policy parameter plumbing for that cache path.
3. Adds sandboxing coverage that keeps `com.apple.trustd.agent` for TLS
while rejecting the cache write rule.
## Why
This closes the exact cache poisoning boundary. The earlier `gh` TLS
issue is now covered by trustd access, so the cache write is no longer
needed.
## Validation
1. Rust formatting passed.
2. The sandboxing crate tests passed.
3. Local macOS Seatbelt repro with patched policy passed. `gh api`
returned `21442` without the cache write rule.
Provider initialization installs process-global OTEL state, so invalid
trace metadata needs to fail before setup begins.
Use the same span attribute validator as config loading when traces are
exported so provider startup enforces the config contract without
duplicating validation logic.
## Why
Some consumers expect conventional hyphenated HTTP headers. Codex
already sends the session and thread IDs on outbound Responses requests,
but it only uses the underscore spellings today, which makes those IDs
harder to consume in systems that normalize or reject underscore header
names.
Full context here:
https://openai.slack.com/archives/C08KCGLSPSQ/p1778248578422369
## What changed
- `build_session_headers` now emits both `session_id` and `session-id`
when a session ID is present.
- It does the same for `thread_id` and `thread-id`.
- Added regression coverage in `codex-api/tests/clients.rs` and
`core/tests/suite/client.rs` so both the lower-level client tests and
the end-to-end request tests assert the two header spellings are
present.
## Test plan
- Added header assertions in `codex-api/tests/clients.rs`.
- Added request-header assertions in `core/tests/suite/client.rs` for
both the `/v1/responses` and `/api/codex/responses` request paths.
Fixes#21665.
## Why
The TUI status line is the right place for compact, glanceable session
state. The original request was motivated by the need to see the active
permission posture without opening `/permissions` or `/status`,
especially when switching between safer and more permissive modes during
a session.
This PR intentionally separates `permissions` from `approval-mode`
instead of combining them into one status-line item. They answer related
but different questions: `permissions` describes the active
sandbox/profile shape, while `approval-mode` describes how command
approvals are handled. Keeping them separate makes each item
independently configurable and avoids long combined labels in an already
space-constrained status line.
The tradeoff is that users who want the full permission posture in the
status line need to opt into both items. In exchange, users can show
only the sandbox/profile label, only the approval behavior, or both, and
named user-defined profiles remain concise. Non-standard permission
shapes are rendered as `Custom permissions` rather than trying to
squeeze detailed profile contents into the status line; `/status`
remains the fuller explanatory surface.
## What changed
- Added a configurable `permissions` status-line item.
- Added a separate `approval-mode` status-line item, with `approval` as
an alias.
- Render standard permission states compactly as `Read Only`,
`Workspace`, or `Full Access`.
- Preserve user-defined permission profile names directly in the status
line.
- Render unnamed non-standard permission shapes as `Custom permissions`.
- Refresh status surfaces when `/permissions` updates the permission
profile, approval policy, or approval reviewer.
- Updated status-line preview snapshot coverage for the new items.
## Verification
- `cargo test -p codex-tui
status_permissions_non_default_workspace_write_uses_workspace_label`
- `cargo test -p codex-tui
permissions_selection_emits_history_cell_when_selection_changes`
- `cargo insta pending-snapshots --manifest-path tui/Cargo.toml`
## Why
The configurable `/statusline` and terminal title can display session
token usage. That display was using the raw total token count, which
includes cached input tokens, so it significantly overstated the token
usage compared with the blended token count shown elsewhere (in
`/status` and tracked in goals). This inconsistency resulted in user
confusion. We don't want to report cached tokens because we don't charge
for them and they are somewhat of an implementation detail that users
shouldn't care about.
## What changed
- Use `TokenUsage::blended_total()` for the `used-tokens` status surface
item so cached input is excluded.
- Add a brief comment to `tokens_in_context_window()` clarifying that it
returns raw `total_tokens`, whose meaning depends on whether the caller
has last-turn or accumulated usage.
## Summary
- enable `apply_patch_freeform` by default in the feature registry
## Why
- make the freeform `apply_patch` tool available by default when model
metadata does not explicitly opt into another mode
## Validation
- `just fmt`
- did not run tests
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
`service_tier` in `config.toml` and profile config was still modeled as
an enum, which blocked newer or experimental service tier IDs even
though the runtime paths already carry string IDs.
This change makes the TOML-facing config accept string service tier IDs
directly while keeping the legacy `fast` alias behavior by normalizing
it to the request value `priority`.
## What Changed
- change the TOML-facing `service_tier` fields in global and profile
config to `Option<String>`
- keep config-load normalization so legacy `fast` still resolves to
`priority`
- persist resolved service tier strings directly in config locks so
arbitrary IDs round-trip cleanly
- regenerate the config schema and add config coverage for arbitrary
string IDs plus legacy `fast` normalization
## Verification
- added config tests for arbitrary string service tiers and legacy
`fast` normalization
- ran `just write-config-schema`
- CI
---------
Co-authored-by: Codex <noreply@openai.com>
## What changed
- rewrote `shutdown_flushes_pending_metadata_irrelevant_updated_at` to
seed an existing pending `updated_at` touch directly in
`RolloutWriterState`
- kept the shutdown test focused on draining a pending touch, leaving
the separate coalescing test to cover timing-based deferral
## Why
The old test had to complete several async operations inside the 50 ms
test-only coalescing window. When that sequence took longer, the second
flush updated `threads.updated_at` immediately and the pre-shutdown
equality assertion failed, even though shutdown behavior was correct.
## Validation
- `cargo test -p codex-rollout
shutdown_flushes_pending_metadata_irrelevant_updated_at`
- `cargo test -p codex-rollout`
Co-authored-by: Codex <noreply@openai.com>
