Commit Graph

9877 Commits

Author SHA1 Message Date
Benjamin Carlsson
d47e5cc0e2 Add Vim buffer jump motions (#40958)
## What changed

- Add `gg` and `G` motions to jump to the first and last buffer lines.
- Support the motions with delete, change, and yank operators, including dot-repeat for edits.
- Expose normal-mode and operator-motion bindings in the configurable keymap while preserving conflicting user bindings.

## Testing

- Cover buffer jumps, operator combinations, repeat behavior, chord routing, and custom remapping.

GitOrigin-RevId: 24a9178cf7964053aaf9a8cd268c5434dda0778e
2026-08-26 19:31:52 +00:00
willwang-openai
6ac012a0d4 Honor layered configuration when loading plugins (#40954)
## What changed

- Resolve plugin activation, MCP server policy, and marketplace definitions from the effective configuration stack, including system settings and trusted project overrides.
- Load plugin skills independently for each working directory passed to `skills/list`.
- Exclude project configuration from catalog requests whose `cwds` are omitted or empty.
- Cache up to eight loaded plugin configurations with least-recently-used eviction so requests for different working directories can reuse their results.

## Testing

- Cover layered plugin and MCP policy across trusted and untrusted projects.
- Cover per-directory skill and plugin catalog results, cache reuse and eviction, and system-defined marketplaces.

GitOrigin-RevId: cb68b2b1bdab1da48e05ca7562bc05014431fc71
2026-08-26 19:02:02 +00:00
Francis Chalissery
7276d67081 Expose resumable misalignment details through app server (#40952)
## Why

App-server clients need the explanation and steering instruction returned with a
misalignment policy violation to offer a user-confirmed continuation. Missing or
incomplete details must continue to behave as a terminal block.

## What changed

- Parse optional misalignment classification, explanation, and steering details
  from streamed and HTTP Responses errors and propagate them through core errors.
- Include the details in live app-server `error` and `turn/completed` payloads and
  export the corresponding protocol schemas and TypeScript types.
- Keep explanations and steering messages out of serialized rollout events and
  redact them from debug output.
- Document how clients can resume with `turn/start` after user confirmation.

## Testing

- Cover streamed, HTTP, and WebSocket-wrapped errors, including malformed and
  classification-only details.
- Verify live app-server notifications expose resumable details without writing
  the explanation or steering message to the rollout.

GitOrigin-RevId: 329258a444c2cd91d0c57ab1720830b33ddcfac5
2026-08-26 18:42:44 +00:00
rka-oai
0e9a2bae52 Enable clock tools for persistent reasoning turns (#40942)
## What changed

- Default the current-time reminder and interruptible `clock.sleep` tool on for
  turns using persistent reasoning effort.
- Preserve explicit `features.current_time_reminder` settings and managed
  requirements, and gate reminder recording on the resolved feature state.

## Testing

- Cover configured, persistent, ordinary-effort, explicitly disabled, and
  managed-disabled clock setups.
- Verify review turns do not expose clock tools or current-time reminders.

GitOrigin-RevId: 4ff8805807ceeef374abd962b1d9e7c37f1fbd83
2026-08-26 18:06:12 +00:00
Steve Coffey
e0c727de04 Classify streaming rate-limit errors (#40931)
## What changed

- Classify `response.failed` events with the `rate_limit_exceeded` code as a distinct retryable error while preserving any parsed retry delay.
- Expose the error as `rateLimitExceeded` through the core protocol and app-server schemas after stream retries are exhausted.
- Preserve the upstream message for TUI display while keeping it out of telemetry summaries.

## Testing

- Cover SSE classification, retry metadata, protocol conversion and serialization, exhausted stream retries, telemetry redaction, and TUI rendering.

GitOrigin-RevId: 02dab4d3477dcd7653a58c49c4bd38687a616579
2026-08-26 17:45:42 +00:00
sayan-oai
7625bd5665 Honor environment-resolved workspace roots (#40912)
## What changed

- Add effective workspace roots to `EnvironmentConfig` and use them when
  materializing permission profiles and filesystem context.
- Preserve selection roots for thread-owned configurations while allowing a
  ready environment attachment to supply its resolved roots.
- Propagate resolved roots to child environments and validate both command and
  `apply_patch` writes in owner-provided secondary workspaces.

GitOrigin-RevId: 1832c168ce1fc37fa67efc2a472998ab4f448c70
2026-08-26 16:28:35 +00:00
jif
5b92c2d2f5 Make Guardian risk score persistence opt-in (#40911)
## What changed

- Add the `features.guardianv2.persist_scores` configuration option for debugging.
- Default the option to `false`, so Guardian V2 reviewed actions and risk scores are only written to rollout files when explicitly enabled and the session is not ephemeral.

## Testing

- Cover configuration deserialization, default non-persistence, and opt-in persistence for nested code-mode actions.

GitOrigin-RevId: 452af99ef14553d45140340ececc2913389124d0
2026-08-26 16:23:34 +00:00
jif
bde9db1375 Record actual Responses endpoints in tracing spans (#40906)
## Why

Responses-compatible requests can use routes other than `/responses`, but their
tracing spans reported the default route regardless of the selected endpoint.

## What changed

- Populate `api.path` from the selected `ResponsesEndpoint` for HTTP and
  WebSocket requests.
- Retain the endpoint on WebSocket connections so stream-request spans report
  the same route as connection spans.

GitOrigin-RevId: 349eca94b27d4cc3a1adde8b257313506bdaf0f4
2026-08-26 15:31:30 +00:00
jif
d4998d611a Record reviewed actions with security risk scores (#40901)
## What changed

- Add optional `call_id` and `action` fields to `SecurityRiskScore` so a persisted score identifies the tool call and bounded action that produced it.
- Populate the fields after Guardian V2 classifies an action while leaving fail-closed and legacy scores without provenance.

## Testing

- Verify scores preserve action details for direct tool calls and nested code-mode calls.
- Verify rollout serialization retains the new fields.

GitOrigin-RevId: fce69af37f368cecf1d9bce42bfc38a969e969b7
2026-08-26 15:02:21 +00:00
jif
62fb56ee56 Route Guardian inference through dedicated endpoints (#40892)
## What changed

- Add the opt-in `features.guardianv2.free_guardian` setting to route eligible
  Guardian reviews to `/guardian` and asynchronous classifications to
  `/guardian-classifier`.
- Support endpoint selection for both HTTP and WebSocket Responses clients,
  including endpoint-aware connection reuse.
- Keep `/responses` for API-key authentication, custom provider URLs,
  non-Guardian models, and configurations where the option is disabled.
- Omit routing hints and `service_tier` from requests sent to the dedicated
  Guardian endpoints.

## Testing

- Cover route selection across authentication modes, provider URLs, HTTP
  fallback, WebSocket review sessions, and classifier sampling.

GitOrigin-RevId: 9c88673c710d009e0239559b9245a4e6c4f4befc
2026-08-26 14:36:48 +00:00
jif
10d5a603ae Persist Guardian V2 risk scores without restoring them (#40884)
## What changed

- Append accepted Guardian V2 classification results to rollout history for non-ephemeral threads.
- Keep resumed and forked threads from restoring a persisted score into active Guardian state.

## Testing

- Verify that asynchronous scoring records the resulting `SecurityRiskScore` rollout item.
- Seed resume and fork tests with a persisted score and verify that approvals ignore it.

GitOrigin-RevId: 40d63a2ea7e7cf6398402c6aed7d5e5727dc9d68
2026-08-26 13:54:36 +00:00
jif
f5420174da Include originating item IDs in MCP request metadata (#40866)
## What changed

- Add the originating Responses item ID to MCP tool request metadata as
  `_meta.itemId`.
- Preserve the initiating `exec` item ID across code-mode cells so nested MCP
  calls carry the same origin.

## Testing

- Cover metadata generation for direct and code-mode MCP tool calls.

GitOrigin-RevId: 1161cac64f1cc3a6d4dac4368e63804bd9d2c417
2026-08-26 10:39:05 +00:00
zsol-openai
74772623db Stop overriding Rust debuginfo in Bazel CI (#40864)
## What changed

Remove the CI-specific `-Cdebuginfo=0` flags for both target and execution
Rust actions, allowing the Bazel Rust toolchain and compilation mode to select
the debug-info level.

GitOrigin-RevId: f785347cde2b3ad0b7814ece2ba44fd89aa823b3
2026-08-26 10:32:57 +00:00
jif
039eb58a0b Give Guardian reviewers read-only access to parent history (#40848)
## Why

Guardian approval reviews may need the original user instructions to decide
whether a command is authorized.

## What changed

- Forward the parent thread's `history.list_windows`, `history.list_items`,
  `history.read_item`, and `history.search_contents` executors to Guardian
  reviewer sessions.
- Keep other parent extension tools, including the `notes` namespace, isolated
  from Guardian reviewers.

## Testing

Add an app-server integration test that reads parent history during a Guardian
review and verifies that `notes` tools are not exposed.

GitOrigin-RevId: 65a90036577050adfa23b1d98f862233c994d9d7
2026-08-26 09:23:41 +00:00
olliem-oai
9dea1f709f Default Guardian v2 to computer-use reviews with images (#40846)
## What changed

- Default Guardian v2's review scope to computer-use tools. Set
  `features.guardianv2.review_scope.computer_use_only = false` to retain the
  broader tool review scope.
- Include images in Guardian transcripts by default. Set
  `features.guardianv2.transcript.include_images = false` to disable them.

## Testing

- Update configuration and app-server coverage for the new defaults and
  explicit broader-scope overrides.

GitOrigin-RevId: 88c3cfe6dd2a96df052111794ad59c5c9ea39b78
2026-08-26 09:13:33 +00:00
olliem-oai
a9ed4f154a Refine Guardian predictive risk classification (#40844)
## What changed

- Evaluate recent unknown-authorization or critical-risk actions over the last five actions instead of ten.
- Flag predicted high- or critical-risk actions when they are credibly expected within the next two actions, rather than treating any potential risk within ten actions as high.
- Remove privileged or broad access as a standalone high-risk condition and classify a trajectory as low when none of the explicit high-risk conditions apply.
- Clarify the user-authorization and reviewer terminology in the classifier instructions.

GitOrigin-RevId: 1eaf63c3327f57a9c8560df085e1fe0afad99cd4
2026-08-26 09:02:09 +00:00
Abhinav
a26f1806a4 Use issuing step settings for approval decisions (#40821)
## Why

Tool actions and approval retries can outlive the step that issued them. If approval handling reads newer turn settings, a delayed action can use the wrong approval policy or reviewer.

## What changed

- Capture the approval policy and reviewer in `GuardianReviewContext` from the issuing step.
- Use step settings for command, patch, permission, sandbox-retry, and remote-network approval decisions.
- Use the active task's published settings for background network approvals and long-lived Unix shell interception.
- Preserve explicit MCP reviewer overrides.

## Testing

Added regression coverage for permission requests, same-turn network policy changes, delayed commands after a new turn starts, sandbox retries, patch policy selection, and MCP reviewer overrides.

GitOrigin-RevId: b7e84a2e1e79a71502272b15b5ef5b0e5643f8b0
2026-08-26 06:51:58 +00:00
zm-oai
21c58c90f2 Harden Windows sandbox helper cleanup (#40808)
## Why

Windows filesystem operations can complete while their helper process remains
alive, and elevated sandbox processes must not survive a disconnected control
pipe.

## What changed

- Read newline-delimited filesystem helper responses without imposing an
  operation deadline, then bound the time spent reaping the helper.
- Drain helper stderr concurrently, retain a bounded diagnostic on failure, and
  kill helpers that do not exit after responding.
- Terminate the elevated runner's job or process when its input pipe closes or
  fails.

## Testing

Added coverage for delayed responses, noisy stderr, stuck helpers, duplicated
file handles, and termination of elevated descendants after a control-pipe
disconnect.

GitOrigin-RevId: c0d3d5d76e13318de17638df4db71fcfb9a7397f
2026-08-26 06:02:56 +00:00
Abhinav
04907ab95a Use step settings for MCP request metadata (#40807)
## Why

MCP tool calls can be issued by a model step whose settings differ from the
turn's initial settings. Request metadata must describe the issuing step.

## What changed

- Build MCP tool-call metadata from the captured `StepContext` model and its
  effective reasoning effort.
- Populate `node_repl_disabled` from the issuing step while leaving turn-owned
  Responses metadata unchanged.
- Apply the same step-scoped values to MCP requests emitted by stop hooks.

## Testing

Add coverage for configured and model-default reasoning effort, model changes,
Node REPL restrictions, and preservation of the original step and turn
metadata.

GitOrigin-RevId: f9cce52fbe559f15437bd19d699b657229c0d145
2026-08-26 05:58:04 +00:00
rka-oai
3e4707b34b Support persistent reasoning effort (#40799)
## What changed

- Add `persistent` to the reasoning-effort protocol and TypeScript SDK types.
- Show model-advertised persistent reasoning in the TUI as “Persistent”.
- Preserve `persistent` in local configuration while sending the Responses API's `disabled` wire value.

## Testing

- Cover parsing and serialization, request translation, remote model requests, the TUI reasoning selector, and TypeScript CLI argument forwarding.

GitOrigin-RevId: 22fc2fba1975db345c12d555ec12803d8dc0abca
2026-08-26 05:00:23 +00:00
Owen Lin
37f4bb94c9 Respect paginated history when stopping background tasks (#40787)
## Why

Stopping a task from the agents overview may need to look up its active turn. Paginated threads should use the turn history endpoint instead of requesting embedded full history.

## What changed

- Read the thread without turns first to determine its history mode.
- Fetch the first turn page for paginated threads when the app server supports it.
- Fall back to embedded turn history for legacy threads and legacy server negotiation.

## Testing

Added coverage for paginated and legacy threads, including fallback after legacy capability negotiation.

GitOrigin-RevId: 44d506b19e32c6b3c27fac11fb748c23b8e47b5e
2026-08-26 04:23:06 +00:00
Benjamin Carlsson
98ee29c7cd Add Vim character find and till motions (#40785)
## What changed

- Add line-local `f`, `F`, `t`, and `T` motions to Vim normal mode.
- Support the motions with `c`, `d`, and `y`, including dot-repeat for changes.
- Expose separate configurable normal-mode and operator-motion bindings, while preserving existing custom keymaps that use the new defaults.
- Keep character targeting aligned to extended grapheme boundaries and skip embedded composer elements.

## Testing

- Cover forward and backward motion and operator ranges, missing and adjacent targets, cancellation, dot-repeat, custom bindings, shifted terminal key events, grapheme clusters, and embedded elements.

GitOrigin-RevId: 2610cf9a13538d9ae9b890c01417ca25316c1d60
2026-08-26 04:18:49 +00:00
Brandon Zhang
25a6e316c8 Make history and notes tools Bridge-compatible (#40775)
## What changed

- Express nullable history and notes arguments with `anyOf` and remove unsupported schema constraints.
- Forward valid JSON object arguments to the backend without enforcing client-side limits.
- Clarify history ID, ordering, cross-agent path, and consistency semantics in the tool descriptions.

## Testing

- Cover forwarding previously restricted argument values for every history and notes tool.
- Verify app-server requests expose the Bridge-compatible schemas and still reject non-object arguments locally.

GitOrigin-RevId: 28809fe9da5099944fd04962b5e71ffda4163eea
2026-08-26 03:46:23 +00:00
sayan-oai
e24190caa9 Use turn environment settings for sandbox execution (#40771)
## Why

Tool execution can use an environment whose sandbox configuration differs from the turn-wide configuration. Sandbox selection and process setup need to follow the environment that owns the tool request.

## What changed

- Source Windows sandbox level, legacy Landlock mode, and Windows private desktop settings from the selected turn environment across tool orchestration, unified exec, patch application, and zsh fork escalation.
- Preserve Windows sandbox level overrides in restorable thread settings and invalidate MCP configuration when that level changes.

## Testing

- Add a resume test that verifies an elevated Windows sandbox override survives thread restoration.

GitOrigin-RevId: 9d8fc46404919f5ae97c9ed712c0932a1f095e36
2026-08-26 03:28:57 +00:00
yqz-openai
3ba7b6941d Add OpenTelemetry tracing to the code-mode host (#40760)
## What changed

- Add `--otel-trace-exporter` for sending OTLP/HTTP JSON traces directly to an exporter.
- Add the mutually exclusive `--otel-trace-listen` option for streaming raw OTLP trace batches over a trace-only WebSocket, backed by a loopback OTLP receiver.
- Instrument host startup, transport requests, and gRPC operations with request, session, execution, cell, wait, and tool-call identifiers where available.
- Flush the trace provider with a bounded timeout during shutdown.

GitOrigin-RevId: 4dcdb2bc3b15166dccd5041fb875caea74f3f6fe
2026-08-26 02:49:51 +00:00
Benjamin Carlsson
2764e83626 Preserve transcript overlay state across updates (#40751)
## What changed

- Detach and restore the live tail when rebuilding transcript renderables after history replacement, consolidation, removal, or highlighting.
- Keep an overlay that is following the bottom pinned to the tail when the composer height changes.
- Update cached content height after prepending history so subsequent scroll calculations use the rebuilt content.

## Testing

- Cover history rebuilds with absent, empty, and active live tails.
- Cover bottom-following across terminal height changes and subsequent committed output.

GitOrigin-RevId: be4bf4d0676261cc52864da9acb567573e34a799
2026-08-26 01:20:40 +00:00
andrewgu-oai
00b7152a60 Fix MCP denial assertion for structured output (#40748)
## What changed

Update the attachment-owned permissions test to read the denial message from
the text content item in the restricted MCP tool's structured output.

GitOrigin-RevId: a4b219b6f14f1b40e8c586147f535748b1770b13
2026-08-26 01:02:52 +00:00
Won Park
dc08ace782 Prepare isolated Guardian reviewer sessions (#40742)
## What changed

- Add a policy prompt and output contract for synchronous Guardian reviews.
- Build reviewer thread options with a preferred review model, low reasoning effort when supported, and a fallback to the parent model and effective effort.
- Isolate reviewer sessions from parent instructions, MCP servers, optional features, and write access while preserving read restrictions, resolved environments, managed network constraints, and rollout persistence.
- Mark prepared threads as internal Guardian review sessions for later host-managed startup.

## Testing

- Cover reviewer isolation, read-only environment inheritance, review-model overrides, and parent-model fallback behavior.

GitOrigin-RevId: 5133be99674b64e06e19e7f2dbc86439e1678790
2026-08-26 00:19:33 +00:00
Nick Steele
9b4a0f8a0a Add enterprise IdP identity resolution for MCP OAuth (#40739)
## What changed

- Resolve stored enterprise IdP sessions against discovered authorization
  metadata, requiring the configured issuer, public-client authentication, and
  supported ID-JAG token exchange capabilities.
- Bind OIDC identity claims and MCP resource indicators to their configured
  issuer, client, and server, and require reauthentication when pinned keyring
  credentials are removed or replaced.
- Hold the credential lock while rereading refresh tokens, and isolate
  enterprise credentials by reserved namespace and Codex home.

## Testing

- Cover metadata and claim validation, resource binding, credential replacement
  and keyring failures, refresh locking, expired ID tokens, and credential
  namespace isolation.

GitOrigin-RevId: edce3c6159f7d6831edf72e9608b3fc3f5823c83
2026-08-25 23:51:57 +00:00
pakrym-oai
75cb7c903d Preserve MCP tool output as content items (#40737)
## What changed

- Convert unstructured MCP results into typed function-call output items instead of serializing the entire content array as a text string.
- Keep structured MCP results as serialized text and preserve media, encrypted content, and unknown content through their existing item conversions.
- Drop empty text items during output truncation so they do not consume API array slots.

## Testing

- Cover text-only, mixed unstructured, structured, and image-sanitized MCP results.
- Verify empty text items are discarded under byte- and token-based truncation policies.

GitOrigin-RevId: fa1c5b7dee6f003a094265379dcabdd060386a48
2026-08-25 23:42:50 +00:00
Adam Perry @ OpenAI
62aacbb2c9 Run exec-server compatibility tests under Bazel (#40736)
## What changed

- Add a Bazel test rule that runs the shared Noise relay compatibility suite with current or packaged Codex binaries.
- Cover both app-server/exec-server directions for the current build, release `0.149.1`, and the minimum supported release `0.145.0` on Linux x86-64.
- Remove the standalone version-skew download scripts and the protocol constant they used.

## Testing

The new `exec-server-current-version-test`, `exec-server-stable-release-test`, and `exec-server-minimum-release-test` targets exercise the compatibility suite.

GitOrigin-RevId: 6b3b91eea8e05a38bec1fac128afc54f0d654014
2026-08-25 23:38:49 +00:00
felixxia-oai
02c9f83f74 Use the model info accessor for skill telemetry (#40735)
Update explicit and implicit skill invocation telemetry to read the model slug
through `TurnContext::model_info()`.

GitOrigin-RevId: 24a69d36d03cdd6c692ca38191c668db10573e06
2026-08-25 23:34:03 +00:00
sayan-oai
4213b38f3c Honor attachment-owned permissions for MCP servers (#40728)
## Why

MCP servers attached to executor environments must retain their owner's
permission profile instead of inheriting the thread-wide sandbox authority.

## What changed

- Resolve and capture a permission profile for each enabled MCP server when
  publishing the runtime, and reject calls or elicitations when that authority
  is unavailable.
- Use the captured server profile for tool approval decisions, elicitation, and
  sandbox metadata, including after runtime refreshes.
- Materialize `:workspace_roots` from `PathUri` values so permissions preserve
  the path convention of remote executor environments.
- Give threadless app discovery and resource reads an explicit default
  permission profile.

## Testing

Added coverage for per-server elicitation authority, unresolved attachments,
runtime refreshes, restricted tool calls, and foreign-platform workspace roots.

GitOrigin-RevId: 6b188d4b08b29e9971ff7aa68a7785a0cdf9e394
2026-08-25 22:50:49 +00:00
Alexi Christakis
0b94751cc4 Add telemetry for SQLite log persistence (#40726)
## Why

SQLite log persistence needs visibility into batch size, write latency,
failures, and entries dropped before they reach the writer. Exporter
diagnostics must not feed back into the SQLite log sink and keep metric
exports active indefinitely.

## What changed

- Record write count, duration, batch bytes, entry count, largest-entry size,
  and queue drops with bounded outcome tags.
- Add explicit byte histogram boundaries for log batch metrics.
- Filter OTLP transport noise and the benign unsolicited HTTP/2 PING warning
  from SQLite log persistence to prevent export cycles while retaining other
  HTTP/2 warnings.

## Testing

- Cover metric values, tags, queue-drop reasons, and histogram boundaries.
- Verify successful and failed OTLP HTTP and gRPC exports become idle instead
  of generating recurring SQLite log-write metrics.

GitOrigin-RevId: a63224ae15945fe84adc977a8b7d55bb47a4de53
2026-08-25 22:46:49 +00:00
jwang-openai
c51e7b3736 Add plugin-attributed skill telemetry (#40724)
## What changed

- Add `plugin_id`, `model_slug`, and `reasoning_effort` dimensions to
  `codex.skill.injected` metrics for explicit and implicit skill invocations.
- Propagate plugin IDs from orchestrator skill metadata so resource-backed skill
  invocations can emit the same attribution.
- Record `codex.skill.turn.duration_seconds` once per plugin used in a turn,
  tagged with the model, reasoning effort, and completed, aborted, or error
  status.

GitOrigin-RevId: d249a9c4e85682b1106ac49d9e598535c6cbb06d
2026-08-25 22:29:14 +00:00
Eric Traut
62bfa41a83 Update the biology Trusted Access link (#40723)
## What changed

- Point the biology safety notice's Trusted Access link to the new ChatGPT route.
- Update the TUI rendering snapshots for the new destination.

GitOrigin-RevId: bba3adac8ac492f5407680bd54dbb727409e9496
2026-08-25 22:19:19 +00:00
Nick Steele
f6805328c4 Add enterprise ID-JAG exchange for MCP OAuth (#40722)
## What changed

- Add a non-interactive two-step exchange that obtains an ID-JAG from an
  enterprise identity provider and trades it for a resource-bound MCP bearer
  token.
- Validate trusted endpoint URLs, request inputs, ID-JAG claims, resource and
  scope bindings, and token responses before credentials are forwarded or a
  bearer token is returned.
- Expose structured authentication failures while redacting credentials and
  provider-controlled error details from diagnostics.

## Testing

- Cover successful exchanges, signed scope narrowing, invalid claims and token
  responses, request validation, and error redaction.

GitOrigin-RevId: d716e0e1c2dc6b230cecbc0e9cc09afeee81d599
2026-08-25 22:12:41 +00:00
Charlie Marsh
d6174a8798 Preserve composer hyperlinks across wrapped lines (#40720)
## What changed

- Detect visible HTTP(S) URLs in the composer and attach the complete OSC 8 destination to every wrapped fragment, including when part of the URL is scrolled offscreen.
- Cache hyperlink metadata with the textarea wrap state while leaving masked input and destinations larger than 8 KiB unannotated.
- Make shared hyperlink mapping grapheme-aware, recognize Unicode whitespace boundaries, and scan URLs and trailing delimiters incrementally for large inputs.

## Testing

- Add coverage for wrapped and scrolled URLs, joined emoji, Unicode whitespace, preserved highlighting, masked input, cache invalidation, oversized destinations, and drafts containing many URLs.

GitOrigin-RevId: 40facdf2fe79438c91c82bff2d849c342f1ab28a
2026-08-25 22:05:48 +00:00
Channing Conger
42624fd63b Preserve bounds in reserved tool schemas (#40719)
## Why

Reserved tool parameter constraints must survive schema parsing so the model receives the declared limits.

## What changed

- Preserve `minimum` and `maximum` as exact JSON numbers.
- Preserve `maxLength` and infer a string schema when it is the only type hint.

## Testing

- Cover bounded integer, number, and string schemas through both parsing paths.
- Verify history and notes tools retain their constraints in app-server model requests.

GitOrigin-RevId: 4cc89722b5b226d595b57232ec3b8e825863faf3
2026-08-25 22:01:26 +00:00
Adam Perry @ OpenAI
de70ec840f Add Bazel repositories for pinned Codex releases (#40718)
## What changed

- Add a Bazel module extension that downloads checksum-pinned Linux x86-64 Codex release packages, with the Codex release host and GitHub Releases as sources.
- Generate `codex` and `package` filegroups from each package's `codex-package.json` manifest so consumers can address the entrypoint or the complete package contents.
- Register repositories for Codex `0.145.0` and `0.149.1` in `MODULE.bazel`.

GitOrigin-RevId: a73225bae76cf01a588c7e1b570f3fdd9f6e12ab
2026-08-25 21:56:05 +00:00
felixxia-oai
32fd056313 Support sandboxed exec-server test environments (#40717)
## What changed

- Add a `TestCodexBuilder` entry point that accepts an explicit `TestEnv`.
- Teach the exec-server test fixture to dispatch filesystem-helper invocations.
- Allow the fixture to receive the Linux sandbox executable through
  `CODEX_TEST_LINUX_SANDBOX_EXE` and run the server on a multithreaded Tokio
  runtime.

GitOrigin-RevId: cb9c27e9db0363d4022a084bc186ccc749e25a95
2026-08-25 21:49:59 +00:00
Benjamin Carlsson
23cedf4802 Add thread ownership metadata for managed worktrees (#40716)
## What changed

- Add `WorktreeManager` APIs to bind a managed linked worktree to a thread and read its owner.
- Store the versioned `codex-thread.json` record in Git metadata with atomic, no-clobber writes, making repeat bindings idempotent while rejecting conflicting owners.
- Validate the managed worktree layout and reject primary, nested, and unmanaged checkouts before accessing ownership metadata.
- Run Git metadata queries with repository-selecting environment variables removed and hooks, filesystem monitoring, attributes, and LFS smudging disabled.

## Testing

- Add integration coverage for rejecting primary and unmanaged worktrees and for writing and reading the expected ownership schema.

GitOrigin-RevId: afec9a0ab711fe4017074bcec5807d860e9c88d0
2026-08-25 21:44:43 +00:00
Chris Dong
346c4db7c2 Track image generation request IDs in analytics (#40714)
## What changed

- Read `x-codex-imagegen-request-id` from image generation and edit responses.
- Propagate the ID to `codex_image_generation_event` analytics.
- Keep the ID in process only, excluding it from extension item serialization, generated types, app-server wire data, and rollout history.

## Testing

- Cover response-header extraction and missing-header behavior in the images client.
- Verify end-to-end analytics emission and the unchanged image-generation item wire shape.

GitOrigin-RevId: 9b8e09b7d655c0c2faa84f0eee99aabae583668d
2026-08-25 21:38:10 +00:00
Adam Perry @ OpenAI
1bc02aea50 Sanitize credentials from Git remote metadata (#40713)
## Why

Git remote URLs can embed usernames, passwords, or tokens. Codex carries these
URLs through turn metadata and persisted thread metadata, so credentials must be
removed before a remote enters those paths.

## What changed

- Add `SanitizedGitUrl`, which parses Git URL and SCP-style remotes, strips
  authentication data, and preserves the conventional `git` SSH user.
- Use sanitized remotes when collecting Git metadata, enriching model requests,
  discovering cloud environments, updating thread metadata, and reading legacy
  rollouts. Reject or omit malformed remotes instead of retaining their raw
  contents.
- Keep API, schema, and TypeScript representations as strings while enforcing
  sanitization in Rust.

## Testing

- Cover URL schemes, SCP and IPv6 forms, remote helpers, encoded paths,
  malformed values, and legacy deserialization.
- Verify credentials do not appear in model requests, API responses, SQLite, or
  rollout files.

GitOrigin-RevId: 6435efc4c45bfbfad4723ce7a0457cb00175f25c
2026-08-25 21:34:05 +00:00
Adam Perry @ OpenAI
399be2d6b5 Move relay helpers into exec-server test support (#40712)
## What changed

- Move the reusable WebSocket relay, registration, frame capture, and encryption assertion helpers into `codex-exec-server-test-support`.
- Re-export the helpers from the existing relay test module so its callers keep the same interface.
- Add the Cargo dependencies and Bazel proto compile data required by the shared support module.

GitOrigin-RevId: 0383149c79ef86cb5e64a09baec41332f9eec4c6
2026-08-25 21:31:40 +00:00
Rasmus Rygaard
eb49f491c6 Add explicit remote executor connection refresh (#40710)
## Why

Planned executor replacement needs a fresh session without waiting for the old
session's transient-disconnect recovery to finish.

## What changed

- Add `Environment::refresh_connection` for remote Noise registry-backed
  environments. It performs a fresh registry lookup, reuses a healthy session
  when the executor identity is unchanged, and connects to a replacement when
  it has changed.
- Retire superseded sessions and connection attempts so they cannot publish
  stale state, accept late RPC results, or replay outstanding work.
- Preserve the existing environment and filesystem handles while replacing the
  underlying client, and require a live status probe before refresh succeeds.

## Testing

Add coverage for replacement and session reuse, recovery and connection races,
lookup and handshake failures, handle preservation, and late RPC responses.

GitOrigin-RevId: f1d11208cbfe8af8feb25f6b6b8100da82169a99
2026-08-25 21:23:17 +00:00
vkg-oai
5ca4175295 Rename host instruction payload to Instructions (#40709)
## What changed

Rename the `UserInstructions` type to `Instructions` across the extension API,
its core API re-export, and all consumers. This keeps loading behavior and the
`LoadedUserInstructions` provider interface unchanged.

GitOrigin-RevId: 1ac236b34c31718e7aa38ff5eae5e9b179aeedd8
2026-08-25 21:20:51 +00:00
alexsong-oai
13fe2bcb7a Redact Bedrock API keys from debug output (#40706)
## Why

Derived debug formatting exposed the managed Amazon Bedrock API key.

## What changed

Implement custom `Debug` formatting for `BedrockApiKeyAuth` that replaces the
API key with `<redacted>` while retaining the region.

## Testing

Add regression coverage for direct and `CodexAuth`-wrapped debug output.

GitOrigin-RevId: 68920fa00a9189c959b0025f22797b32f53d1995
2026-08-25 21:08:22 +00:00
Felipe Coury
6988d390b3 Generate automatic and manual conversation recaps (#40705)
## Why

Users returning to a conversation need a short summary of its recent progress and next steps.

## What changed

- Generate a recap for eligible idle, unfocused conversations and add it to the transcript.
- Add `/recap` to request the same summary on demand, with visible progress and actionable errors.
- Keep recap requests bounded and isolated, reject stale results, and retry only automatic failures.

## Testing

- Cover automatic and manual generation, request eligibility, failure handling, loading-state cleanup, and temporary-thread configuration.

GitOrigin-RevId: 926a88d6996d67a42809ce2c4091b13fca882b52
2026-08-25 21:03:11 +00:00
Felipe Coury
40ba7da7b4 Prepare automatic TUI recap generation (#40697)
## What changed

- Schedule recap eligibility checks when an unfocused thread changes, and cancel them when focus returns or recap state is replaced.
- Add bounded structured recap requests with stale-result rejection, normalized output, and one retry per turn revision.
- Harden temporary structured threads by reading the effective configuration and disabling all discovered MCP servers, tools, skills, and environment access.

## Testing

- Cover recap timers, focus changes, failed turns, output bounds, stale results, and retry behavior.

GitOrigin-RevId: eebf50059763928e9a2444b14d51648e0bdff4a8
2026-08-25 20:56:03 +00:00