## Note-- added plugin mentions via @, but that conflicts with file
mentions
depends and builds upon #13433.
- introduces explicit `@plugin` mentions. this injects the plugin's mcp
servers, app names, and skill name format into turn context as a dev
message.
- we do not yet have UI for these mentions, so we currently parse raw
text (as opposed to skills and apps which have UI chips, autocomplete,
etc.) this depends on a `plugins/list` app-server endpoint we can feed
the UI with, which is upcoming
- also annotate mcp and app tool descriptions with the plugin(s) they
come from. this gives the model a first class way of understanding what
tools come from which plugins, which will help implicit invocation.
### Tests
Added and updated tests, unit and integration. Also confirmed locally a
raw `@plugin` injects the dev message, and the model knows about its
apps, mcps, and skills.
## Summary
This updates the `js_repl` prompt and docs to make the image guidance
less confusing.
## What changed
- Clarified that `codex.emitImage(...)` adds one image per call and can
be called multiple times to emit multiple images.
- Reworded the image-encoding guidance to be general `js_repl` advice
instead of `ImageDetailOriginal`-specific behavior.
- Updated the guidance to recommend JPEG at about quality 85 when lossy
compression is acceptable, and PNG when transparency or lossless detail
matters.
- Mirrored the same wording in the public `js_repl` docs.
This improves macOS Seatbelt handling for sandboxed tool processes.
## Changes
- Allow dual-stack local binding in proxy-managed sessions, while still
keeping traffic limited to loopback and configured proxy endpoints.
- Replace the old generic unix-socket path rule with explicit AF_UNIX
permissions for socket creation, bind, and outbound connect.
- Keep explicitly approved wrapper sockets connect-only.
Local helper servers are less likely to fail when binding on macOS.
Tools using local unix-socket IPC should work more reliably under the
sandbox.
Full-network sessions, proxy fail-closed behavior, and proxy lifecycle
are unchanged.
## Summary
- always pass `--unshare-user` in the Linux bubblewrap argv builders
- stop relying on bubblewrap's auto-userns behavior, which is skipped
for `uid 0`
- update argv expectations in tests and document the explicit user
namespace behavior
The installed Codex binary reproduced the same issue with:
- `codex -c features.use_linux_sandbox_bwrap=true sandbox linux -- true`
- `bwrap: Creating new namespace failed: Operation not permitted`
This happens because Codex asked bubblewrap for mount/pid/network
namespaces without explicitly asking for a user namespace. In a
root-inside-container environment without ambient `CAP_SYS_ADMIN`, that
fails. Adding `--unshare-user` makes bubblewrap create the user
namespace first and then the remaining namespaces succeed.
This PR adds a durable trace linkage for each turn by storing the active
trace ID on the rollout TurnContext record stored in session rollout
files.
Before this change, we propagated trace context at runtime but didn’t
persist a stable per-turn trace key in rollout history. That made
after-the-fact debugging harder (for example, mapping a historical turn
to the corresponding trace in datadog). This sets us up for much easier
debugging in the future.
### What changed
- Added an optional `trace_id` to TurnContextItem (rollout schema).
- Added a small OTEL helper to read the current span trace ID.
- Captured `trace_id` when creating `TurnContext` and included it in
`to_turn_context_item()`.
- Updated tests and fixtures that construct TurnContextItem so
older/no-trace cases still work.
### Why this approach
TurnContext is already the canonical durable per-turn metadata in
rollout. This keeps ownership clean: trace linkage lives with other
persisted turn metadata.
### Motivation
- Prevent untrusted js_repl code from supplying arbitrary external URLs
that the host would forward into model input and cause external fetches
/ data exfiltration. This change narrows the emitImage contract to safe,
self-contained data URLs.
### Description
- Kernel: added `normalizeEmitImageUrl` and enforce that string-valued
`codex.emitImage(...)` inputs and `input_image`/content-item paths only
accept non-empty `data:` URLs; byte-based paths still produce data URLs
as before (`kernel.js`).
- Host: added `validate_emitted_image_url` and check `EmitImage`
requests before creating `FunctionCallOutputContentItem::InputImage`,
returning an error to the kernel if the URL is not a `data:` URL
(`mod.rs`).
- Tests/docs: added a runtime test
`js_repl_emit_image_rejects_non_data_url` to assert rejection of
non-data URLs and updated user-facing docs/instruction text to state
`data URL` support instead of generic direct image URLs (`mod.rs`,
`docs/js_repl.md`, `project_doc.rs`).
### Testing
- Ran `just fmt` in `codex-rs`; it completed successfully.
- Added a runtime test (`cargo test -p codex-core
js_repl_emit_image_rejects_non_data_url`) but executing the test in this
environment failed due to a missing system dependency required by
`codex-linux-sandbox` (the vendored `bubblewrap` build requires
`libcap.pc` via `pkg-config`), so the test could not be run here.
- Attempted a focused `cargo test` invocation with and without default
features; both compile/test attempts were blocked by the same missing
system `libcap` dependency in this environment.
------
[Codex
Task](https://chatgpt.com/codex/tasks/task_i_69a7837bce98832d91db92d5f76d6cbe)
## Summary
This changes the Unix shell escalation path for skill-matched
executables to apply a skill's `PermissionProfile` as additive
permissions on top of the existing turn/request sandbox policy.
Previously, skill-matched executables compiled the skill permission
profile into a standalone sandbox policy and executed against that
replacement policy. Now they go through the same
`additional_permissions` merge path used elsewhere in shell sandbox
preparation.
## What Changed
- Changed `skill_escalation_execution()` to return
`EscalationPermissions::PermissionProfile(...)` for non-empty skill
permission profiles.
- Kept empty or missing skill permission profiles on the `TurnDefault`
path.
- Added tests covering the new additive skill-permission behavior.
- Added inline comments in `prepare_escalated_exec()` clarifying the
difference between additive permission merging and fully specified
replacement sandbox policies.
- Removed the now-unused skill permission compiler module after
switching this path away from standalone compiled skill sandbox
policies.
## Testing
- Ran `just fmt` in `codex-rs`
- Ran `cargo test -p codex-core`
`cargo test -p codex-core` still hits an unrelated existing failure:
`shell_snapshot::tests::snapshot_shell_does_not_inherit_stdin`
## Follow-up
This change intentionally does not merge skill-specific macOS seatbelt
profile extensions through the `additional_permissions` path yet.
Filesystem and network permissions now follow the additive merge path,
but seatbelt extension permissions still need separate handling in a
follow-up PR.
## Summary
- Change `js_repl` failed-cell persistence so later cells keep prior
bindings plus only the current-cell bindings whose initialization
definitely completed before the throw.
- Preserve initialized lexical bindings across failed cells via
module-namespace readability, including top-level destructuring that
partially succeeds before a later throw.
- Preserve hoisted `var` and `function` bindings only when execution
clearly reached their declaration site, and preserve direct top-level
pre-declaration `var` writes and updates through explicit write-site
markers.
- Preserve top-level `for...in` / `for...of` `var` bindings when the
loop body executes at least once, using a first-iteration guard to avoid
per-iteration bookkeeping overhead.
- Keep prior module state intact across link-time failures and
evaluation failures before the prelude runs, while still allowing failed
cells that already recreated prior bindings to persist updates to those
existing bindings.
- Hide internal commit hooks from user `js_repl` code after the prelude
aliases them, so snippets cannot spoof committed bindings by calling the
raw `import.meta` hooks directly.
- Add focused regression coverage for the supported failed-cell
behaviors and the intentionally unsupported boundaries.
- Update `js_repl` docs and generated instructions to describe the new,
narrower failed-cell persistence model.
## Motivation
We saw `js_repl` drop bindings that had already been initialized
successfully when a later statement in the same cell threw, for example:
const { context: liveContext, session } =
await initializeGoogleSheetsLiveForTab(tab);
// later statement throws
That was surprising in practice because successful earlier work
disappeared from the next cell.
This change makes failed-cell persistence more useful without trying to
model every possible partially executed JavaScript edge case. The
resulting behavior is narrower and easier to reason about:
- prior bindings are always preserved
- lexical bindings persist when their initialization completed before
the throw
- hoisted `var` / `function` bindings persist only when execution
clearly reached their declaration or a supported top-level `var` write
site
- failed cells that already recreated prior bindings can persist writes
to those existing bindings even if they introduce no new bindings
The detailed edge-case matrix stays in `docs/js_repl.md`. The
model-facing `project_doc` guidance is intentionally shorter and focused
on generation-relevant behavior.
## Supported Failed-Cell Behavior
- Prior bindings remain available after a failed cell.
- Initialized lexical bindings remain available after a failed cell.
- Top-level destructuring like `const { a, b } = ...` preserves names
whose initialization completed before a later throw.
- Hoisted `function` bindings persist when execution reached the
declaration statement before the throw.
- Direct top-level pre-declaration `var` writes and updates persist, for
example:
- `x = 1`
- `x += 1`
- `x++`
- short-circuiting logical assignments only persist when the write
branch actually runs
- Non-empty top-level `for...in` / `for...of` `var` loops persist their
loop bindings.
- Failed cells can persist updates to existing carried bindings after
the prelude has run, even when the cell commits no new bindings.
- Link failures and eval failures before the prelude do not poison
`@prev`.
## Intentionally Unsupported Failed-Cell Cases
- Hoisted function reads before the declaration, such as `foo(); ...;
function foo() {}`
- Aliasing or inference-based recovery from reads before declaration
- Nested writes inside already-instrumented assignment RHS expressions
- Destructuring-assignment recovery for hoisted `var`
- Partial `var` destructuring recovery
- Pre-declaration `undefined` reads for hoisted `var`
- Empty top-level `for...in` / `for...of` loop vars
- Nested or scope-sensitive pre-declaration `var` writes outside direct
top-level expression statements
This adds a first-class server request for MCP server elicitations:
`mcpServer/elicitation/request`.
Until now, MCP elicitation requests only showed up as a raw
`codex/event/elicitation_request` event from core. That made it hard for
v2 clients to handle elicitations using the same request/response flow
as other server-driven interactions (like shell and `apply_patch`
tools).
This also updates the underlying MCP elicitation request handling in
core to pass through the full MCP request (including URL and form data)
so we can expose it properly in app-server.
### Why not `item/mcpToolCall/elicitationRequest`?
This is because MCP elicitations are related to MCP servers first, and
only optionally to a specific MCP tool call.
In the MCP protocol, elicitation is a server-to-client capability: the
server sends `elicitation/create`, and the client replies with an
elicitation result. RMCP models it that way as well.
In practice an elicitation is often triggered by an MCP tool call, but
not always.
### What changed
- add `mcpServer/elicitation/request` to the v2 app-server API
- translate core `codex/event/elicitation_request` events into the new
v2 server request
- map client responses back into `Op::ResolveElicitation` so the MCP
server can continue
- update app-server docs and generated protocol schema
- add an end-to-end app-server test that covers the full round trip
through a real RMCP elicitation flow
- The new test exercises a realistic case where an MCP tool call
triggers an elicitation, the app-server emits
mcpServer/elicitation/request, the client accepts it, and the tool call
resumes and completes successfully.
### app-server API flow
- Client starts a thread with `thread/start`.
- Client starts a turn with `turn/start`.
- App-server sends `item/started` for the `mcpToolCall`.
- While that tool call is in progress, app-server sends
`mcpServer/elicitation/request`.
- Client responds to that request with `{ action: "accept" | "decline" |
"cancel" }`.
- App-server sends `serverRequest/resolved`.
- App-server sends `item/completed` for the mcpToolCall.
- App-server sends `turn/completed`.
- If the turn is interrupted while the elicitation is pending,
app-server still sends `serverRequest/resolved` before the turn
finishes.
## Why
`shell_zsh_fork` already provides stronger guarantees around which
executables receive elevated permissions. To reuse that machinery from
unified exec without pushing Unix-specific escalation details through
generic runtime code, the escalation bootstrap and session lifetime
handling need a cleaner boundary.
That boundary also needs to be safe for long-lived sessions: when an
intercepted shell session is closed or pruned, any in-flight approval
workers and any already-approved escalated child they spawned must be
torn down with the session, and the inherited escalation socket must not
leak into unrelated subprocesses.
## What Changed
- Extracted a reusable `EscalationSession` and
`EscalateServer::start_session(...)` in `shell-escalation` so callers
can get the wrapper/socket env overlay and keep the escalation server
alive without immediately running a one-shot command.
- Documented that `EscalationSession::env()` and
`ShellCommandExecutor::run(...)` exchange only that env overlay, which
callers must merge into their own base shell environment.
- Clarified the prepared-exec helper boundary in `core` by naming the
new helper APIs around `ExecRequest`, while keeping the legacy
`execute_env(...)` entrypoints as thin compatibility wrappers for
existing callers that still use the older naming.
- Added a small post-spawn hook on the prepared execution path so the
parent copy of the inheritable escalation socket is closed immediately
after both the existing one-shot shell-command spawn and the
unified-exec spawn.
- Made session teardown explicit with session-scoped cancellation:
dropping an `EscalationSession` or canceling its parent request now
stops intercept workers, and the server-spawned escalated child uses
`kill_on_drop(true)` so teardown cannot orphan an already-approved
child.
- Added `UnifiedExecBackendConfig` plumbing through `ToolsConfig`, a
`shell::zsh_fork_backend` facade, and an opaque unified-exec
spawn-lifecycle hook so unified exec can prepare a wrapped `zsh -c/-lc`
request without storing `EscalationSession` directly in generic
process/runtime code.
- Kept the existing `shell_command` zsh-fork behavior intact on top of
the new bootstrap path. Tool selection is unchanged in this PR: when
`shell_zsh_fork` is enabled, `ShellCommand` still wins over
`exec_command`.
## Verification
- `cargo test -p codex-shell-escalation`
- includes coverage for `start_session_exposes_wrapper_env_overlay`
- includes coverage for `exec_closes_parent_socket_after_shell_spawn`
- includes coverage for
`dropping_session_aborts_intercept_workers_and_kills_spawned_child`
- `cargo test -p codex-core
shell_zsh_fork_prefers_shell_command_over_unified_exec`
- `cargo test -p codex-core --test all
shell_zsh_fork_prompts_for_skill_script_execution`
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/13392).
* #13432
* __->__ #13392
- add a speed row to the startup/session header under the model row
- render the speed row with the same styling pattern as the model row,
using /fast to change
- show only Fast or Standard to users and update the affected snapshots
---------
Co-authored-by: Codex <noreply@openai.com>
add `web_search_tool_type` on model_info that can be populated from
backend. will be used to filter which models can use `web_search` with
images and which cant.
added small unit test.
- lower `submission_dispatch` span logging to debug for realtime audio
submissions only
- keep other submission spans at info and add a targeted test for the
level selection
---------
Co-authored-by: Codex <noreply@openai.com>
## Summary
- add `js_repl` support for dynamic imports of relative and absolute
local ESM `.js` / `.mjs` files
- keep bare package imports on the native Node path and resolved from
REPL-global search roots (`CODEX_JS_REPL_NODE_MODULE_DIRS`, then `cwd`),
even when they originate from imported local files
- restrict static imports inside imported local files to other local
relative/absolute `.js` / `.mjs` files, and surface a clear error for
unsupported top-level static imports in the REPL cell
- run imported local files inside the REPL VM context so they can access
`codex.tmpDir`, `codex.tool`, captured `console`, and Node-like
`import.meta` helpers
- reload local files between execs so later `await import("./file.js")`
calls pick up edits and fixed failures, while preserving package/builtin
caching and persistent top-level REPL bindings
- make `import.meta.resolve()` self-consistent by allowing the returned
`file://...` URLs to round-trip through `await import(...)`
- update both public and injected `js_repl` docs to clarify the narrowed
contract, including global bare-import resolution behavior for local
absolute files
## Testing
- `cargo test -p codex-core js_repl_`
- built codex binary and verified behavior
---------
Co-authored-by: Codex <noreply@openai.com>
- rotate the paid-plan startup promo slot 50/50 between the existing
Codex App promo and a new Fast mode promo
- keep the Fast mode call to action platform-neutral so Windows can show
the same tip
- add a focused unit test to ensure the paid promo pool actually rotates
---------
Co-authored-by: Codex <noreply@openai.com>
### first half of changes, followed by #13510
Track plugin capabilities as derived summaries on `PluginLoadOutcome`
for enabled plugins with at least one skill/app/mcp.
Also add `Plugins` section to `user_instructions` injected on session
start. These introduce the plugins concept and list enabled plugins, but
do NOT currently include paths to enabled plugins or details on what
apps/mcps the plugins contain (current plan is to inject this on
@-mention). that can be adjusted in a follow up and based on evals.
### tests
Added/updated tests, confirmed locally that new `Plugins` section +
currently enabled plugins show up in `user_instructions`.
## Summary
- ensure `thread.resume` reuses the stored `gitInfo` instead of
rebuilding it from the live working tree
- persist and apply thread git metadata through the resume flow and add
a regression test covering branch mismatch cases
## Testing
- Not run (not requested)
Improve observability of realtime conversation event handling by logging
non-audio events with payload details in the event loop, while skipping
audio-out events to reduce noise.
Support marketplace.json that points to a local file, with
```
"source":
{
"source": "local",
"path": "./plugin-1"
},
```
Add a new plugin/install endpoint which add the plugin to the cache folder and enable it in config.toml.
Addresses #13478
Summary
- Add two new scopes for `tui.notifications` config: `plan-mode-prompt`
and `user-input-requested`.
- Add Plan Mode prompt and user-input-requested notifications to the TUI
so these events surface consistently outside of plan mode
- Add helpers and tests to ensure the new notification types publish the
right titles, summaries, and type tags for filtering
- Add prioritization mechanism to fix an existing bug where one
notification event could arbitrarily overwrite others
Testing
- Manually tested plan mode to ensure that notification appeared
### Overview
This PR:
- Updates `app-server-test-client` to load OTEL settings from
`$CODEX_HOME/config.toml` and initializes its own OTEL provider.
- Add real client root spans to app-server test client traces.
This updates `codex-app-server-test-client` so its Datadog traces
reflect the full client-driven flow instead of a set of server spans
stitched together under a synthetic parent.
Before this change, the test client generated a fake `traceparent` once
and reused it for every JSON-RPC request. That kept the requests in one
trace, but there was no real client span at the top, so Datadog ended up
showing the sequence in a slightly misleading way, where all RPCs were
anchored under `initialize`.
Now the test client:
- loads OTEL settings from the normal Codex config path, including
`$CODEX_HOME/config.toml` and existing --config overrides
- initializes tracing the same way other Codex binaries do when trace
export is enabled
- creates a real client root span for each scripted command
- creates per-request client spans for JSON-RPC methods like
`initialize`, `thread/start`, and `turn/start`
- injects W3C trace context from the current client span into
request.trace instead of reusing a fabricated carrier
This gives us a cleaner trace shape in Datadog:
- one trace URL for the whole scripted flow
- a visible client root span
- proper client/server parent-child relationships for each app-server
request
## Problem
The `ansi`, `base16`, and `base16-256` syntax themes are designed to
emit ANSI palette colors so that highlighted code respects the user's
terminal color scheme. Syntect encodes this intent in the alpha channel
of its `Color` struct — a convention shared with `bat` — but
`convert_style` was ignoring it entirely, treating every foreground
color as raw RGB. This caused ANSI-family themes to produce hard-coded
RGB values (e.g. `Rgb(0x02, 0, 0)` instead of `Green`), defeating their
purpose and rendering them as near-invisible dark colors on most
terminals.
Reported in #12890.
## Mental model
Syntect themes use a compact encoding in their `Color` struct:
| `alpha` | Meaning of `r` | Mapped to |
|---------|----------------|-----------|
| `0x00` | ANSI palette index (0–255) | `RtColor::Black`…`Gray` for 0–7,
`Indexed(n)` for 8–255 |
| `0x01` | Unused (sentinel) | `None` — inherit terminal default fg/bg |
| `0xFF` | True RGB red channel | `RtColor::Rgb(r, g, b)` |
| other | Unexpected | `RtColor::Rgb(r, g, b)` (silent fallback) |
This encoding is a bat convention that three bundled themes rely on. The
new `convert_syntect_color` function decodes it; `ansi_palette_color`
maps indices 0–7 to ratatui's named ANSI variants.
| macOS - Dark | macOS - Light | Windows - ansi | Windows - base16 |
|---|---|---|---|
| <img width="1064" height="1205" alt="macos-dark"
src="https://github.com/user-attachments/assets/f03d92fb-b44b-4939-b2b9-503fde133811"
/> | <img width="1073" height="1227" alt="macos-light"
src="https://github.com/user-attachments/assets/2ecb2089-73b5-4676-bed8-e4e6794250b4"
/> |
|
|
## Non-goals
- Background color decoding — we intentionally skip backgrounds to
preserve the terminal's own background. The decoder supports it, but
`convert_style` does not apply it.
- Italic/underline changes — those remain suppressed as before.
- Custom `.tmTheme` support for ANSI encoding — only the bundled themes
use this convention.
## Tradeoffs
- The alpha-channel encoding is an undocumented bat/syntect convention,
not a formal spec. We match bat's behavior exactly, trading formality
for ecosystem compatibility.
- Indices 0–7 are mapped to ratatui's named variants (`Black`, `Red`, …,
`Gray`) rather than `Indexed(0)`…`Indexed(7)`. This lets terminals apply
bold/bright semantics to named colors, which is the expected behavior
for ANSI themes, but means the two representations are not perfectly
round-trippable.
## Architecture
All changes are in `codex-rs/tui/src/render/highlight.rs`, within the
style-conversion layer between syntect and ratatui:
```
syntect::highlighting::Color
└─ convert_syntect_color(color) [NEW — alpha-dispatch]
├─ a=0x00 → ansi_palette_color() [NEW — index→named/indexed]
├─ a=0x01 → None (terminal default)
├─ a=0xFF → Rgb(r,g,b) (standard opaque path)
└─ other → Rgb(r,g,b) (silent fallback)
```
`convert_style` delegates foreground mapping to `convert_syntect_color`
instead of inlining the `Rgb(r,g,b)` conversion. The core highlighter is
refactored into `highlight_to_line_spans_with_theme` (accepts an
explicit theme reference) so tests can highlight against specific themes
without mutating process-global state.
### ANSI-family theme contract
The ANSI-family themes (`ansi`, `base16`, `base16-256`) rely on upstream
alpha-channel encoding from two_face/syntect. We intentionally do
**not** validate this contract at runtime — if the upstream format
changes, the `ansi_themes_use_only_ansi_palette_colors` test catches it
at build time, long before it reaches users. A runtime warning would be
unactionable noise.
### Warning copy cleanup
User-facing warning messages were rewritten for clarity:
- Removed internal jargon ("alpha-encoded ANSI color markers", "RGB
fallback semantics", "persisted override config")
- Dropped "syntax" prefix from "syntax theme" — users just think "theme"
- Downgraded developer-only diagnostics (duplicate override, resolve
fallback) from `warn` to `debug`
## Observability
- The `ansi_themes_use_only_ansi_palette_colors` test enforces the
ANSI-family contract at build time.
- The snapshot test provides a regression tripwire for palette color
output.
- User-facing warnings are limited to actionable issues: unknown theme
names and invalid custom `.tmTheme` files.
## Tests
- **Unit tests for each alpha branch:** `alpha=0x00` with low index
(named color), `alpha=0x00` with high index (`Indexed`), `alpha=0x01`
(terminal default), unexpected alpha (falls back to RGB), ANSI white →
Gray mapping.
- **Integration test:**
`ansi_family_themes_use_terminal_palette_colors_not_rgb` — highlights a
Rust snippet with each ANSI-family theme and asserts zero `Rgb`
foreground colors appear.
- **Snapshot test:** `ansi_family_foreground_palette_snapshot` — records
the exact set of unique foreground colors each ANSI-family theme
produces, guarding against regressions.
- **Warning validation tests:** verify user-facing warnings for missing
custom themes, invalid `.tmTheme` files, and bundled theme resolution.
## Test plan
- [ ] `cargo test -p codex-tui` passes all new and existing tests
- [ ] Select `ansi`, `base16`, or `base16-256` theme and verify code
blocks render with terminal palette colors (not near-black RGB)
- [ ] Select a standard RGB theme (e.g. `dracula`) and verify no
regression in color output
## Summary
- update the /fast slash command description to mention fastest
inference
- mention the 3X plan usage tradeoff in the help copy
## Testing
- cargo test -p codex-tui slash_command (currently blocked by an
unrelated latest-main codex-tui compile error in chatwidget.rs:
refresh_queued_user_messages missing)
---------
Co-authored-by: Codex <noreply@openai.com>
This is PR 3 of the app-server tracing rollout.
PRs https://github.com/openai/codex/pull/13285 and
https://github.com/openai/codex/pull/13368 gave us inbound request spans
in app-server and propagated trace context through Submission. This
change finishes the next piece in core: when a request actually starts a
turn, we now create a core-owned long-lived span that stays open for the
real lifetime of the turn.
What changed:
- `Session::spawn_task` can now optionally create a long-lived turn span
and run the spawned task inside it
- `turn/start` uses that path, so normal turn execution stays under a
single core-owned span after the async handoff
- `review/start` uses the same pattern
- added a unit test that verifies the spawned turn task inherits the
submission dispatch trace ancestry
**Why**
The app-server request span is intentionally short-lived. Once work
crosses into core, we still want one span that covers the actual
execution window until completion or interruption. This keeps that
ownership where it belongs: in the layer that owns the runtime
lifecycle.
The electron app doesn't start up the app-server in a particular
workspace directory.
So sandbox setup happens in the app-installed directory instead of the
project workspace.
This allows the app do specify the workspace cwd so that the sandbox
setup actually sets up the ACLs instead of exiting fast and then having
the first shell command be slow.
Validated login + refresh flows. Removing scopes from the refresh
request until we have upgrade flow in place. Confirmed that tokens
refresh with existing scopes.
Celia Chen ·
sayan-oai ·
Curtis 'Fjord' Hawthorne ·
viyatb-oai ·
Owen Lin ·
rhan-oai ·
Max Johnson ·
jif-oai ·
Michael Bolin ·
pash-openai ·
Ahmed Ibrahim ·
aaronl-openai ·
Matthew Zeng ·
dependabot[bot] ·
joeytrasatti-openai ·
Won Park ·
xl-openai ·
Eric Traut ·
alexsong-oai ·
Felipe Coury ·
Val Kharitonov ·
iceweasel-oai ·
Alex Daley ·