Commit Graph
1171 Commits
Author SHA1 Message Date
efrazer-oai 8e4b7d31de Use jemalloc for Linux musl binaries (#42850)
## What changed

- Configure the CLI and app server to use `tikv-jemallocator` on
  `x86_64-unknown-linux-musl` and `aarch64-unknown-linux-musl`.
- Keep the dependency and global allocator selection scoped to those targets.

GitOrigin-RevId: 93c54c90d9acb7e48edffd032cd8ec685a782210
2026-09-04 20:48:16 +00:00
Felipe Coury 7a8092a447 Preserve Markdown formatting when copying TUI responses (#42847)
## What changed

- Add rendered HTML alongside the original Markdown when copying a whole response through the native clipboard, so rich-text destinations preserve headings, lists, tables, code, and other formatting.
- Keep code-block, blockquote, transcript, terminal, and WSL clipboard copies as plain text.
- Escape raw HTML, omit remote image sources, and render unsupported link destinations as visible text when producing clipboard HTML.

## Testing

- Add coverage for rich formatting, local and unsafe links, inert images, fenced tables, copy selection formats, and clipboard fallbacks.

GitOrigin-RevId: c6f5edd88b65f9e9b1ddc5618442868cf47ee560
2026-09-04 20:38:08 +00:00
iceweasel-oai 60888d0868 Add a native Windows MXC sandbox adapter (#42841)
## What changed

- Add `codex-mxc-sandbox` with native MXC availability detection and a launcher that inherits standard I/O and waits for the sandboxed process.
- Reject unsupported learning-mode and fallback policies, and verify deny-path support before launch.
- Route the existing Windows MXC availability metric through the new crate and move its Windows linking dependencies alongside the adapter.

GitOrigin-RevId: e2a220b85718e00b0e710c30bbe66b6c56958176
2026-09-04 20:09:04 +00:00
ostepanian b3f5e45cc1 Add direct SigV4 transport to exec-server (#42781)
## Why

Allow remote exec servers to connect directly to AWS-hosted registries that
authenticate registry requests and WebSocket handshakes with AWS SigV4.

## What changed

- Add `--remote-transport direct` with SigV4 profile, region, and service
  options while keeping Noise as the default transport.
- Register the `direct_jsonrpc_v1` transport and carry plain exec-server
  JSON-RPC messages over the authenticated WebSocket.
- Reuse direct registrations across transient disconnects, refresh them after
  a `409 Conflict`, and require TLS for non-loopback endpoints.

## Testing

- Cover CLI validation and SigV4 request signing.
- Exercise direct registration, handshake retry behavior, JSON-RPC
  interoperability, and process recovery after reconnecting.

GitOrigin-RevId: 0755df330ba3abe5db0a516fdaa49338d9bbe2d2
2026-09-04 14:49:46 +00:00
Benjamin Carlsson 1b53f6a44e Add WebRTC negotiation to the voice host (#42676)
## What changed

- Add helper protocol requests to create a WebRTC offer, apply an answer, and report readiness once the ordered `oai-events` data channel opens.
- Support UDP and TCP candidates while bounding signaling size, remote candidates, inbound TCP streams, and WebRTC buffers. Redact session descriptions and peer errors from diagnostics.
- Close unexpected remote data channels and tear down the peer during helper shutdown, including when the parent pipe is lost during startup.

## Testing

- Exercise offer/answer negotiation and clean shutdown over real UDP and TCP sockets.
- Cover signaling validation, candidate and connection limits, early packet loss, slow TCP connection setup, and blocked-startup parent loss.

GitOrigin-RevId: 5f5e60a3a4d1d8eaca2e4724e66284f54061b6af
2026-09-04 05:07:02 +00:00
Benjamin Carlsson eb5a00b068 Add managed worktrees to codex exec (#42652)
## What changed

- Add the experimental `worktrees` feature and a shared `--worktree` flag for new and forked `codex exec` sessions.
- Create each enabled session in a managed Git worktree, use that checkout as the session working directory, and bind the checkout to the new thread.
- Share the configured worktree pool with Desktop while leaving automatic cleanup disabled for CLI allocations.
- Reject unsupported commands, remote execution, ignored user configuration, ephemeral sessions, and use without the feature enabled before allocating a worktree.

## Testing

- Cover flag placement and inheritance, supported and rejected command combinations, worktree allocation and thread ownership, configuration gating, and compatibility with existing worktree-backed sessions.

GitOrigin-RevId: 011ff4639b09e8992c50d7b823df23e71798670e
2026-09-04 01:56:38 +00:00
Krish Chainani 03467026f2 Add an injectable attachment store to ThreadManager (#42634)
## What changed

- Add the `codex-attachment-store` crate with storage-neutral attachment metadata, references, errors, and an asynchronous persistence interface.
- Provide an inline implementation that preserves attachment bytes as media-typed base64 data URLs.
- Inject the store into `ThreadManager`, expose it to consumers, and retain inline storage as the default for existing entry points.

## Testing

- Verify inline storage round-trips binary, text, PNG, and JPEG data.
- Verify attachment debug output redacts URLs while retaining file IDs.

GitOrigin-RevId: 7688dcd3c89d7540ed2398f3e7c63881fbfcda97
2026-09-03 23:20:54 +00:00
Benjamin Carlsson d979df154c Initialize the packaged GStreamer runtime in the voice host (#42631)
## Why

The voice helper previously supported only handshake and shutdown, without
validating that its packaged native runtime could be initialized safely.

## What changed

- Add an `initializeRuntime` protocol exchange that loads GStreamer and the
  required plugins from physical package paths without opening audio devices.
- Restrict plugin discovery, registry access, and native library search paths,
  and keep loaded libraries alive until the helper exits.
- Give initialization a dedicated deadline and terminate the owned helper when
  initialization is cancelled or fails.
- Allow binary-only Rust targets to disable the default Bazel library target.

## Testing

- Cover helper-only packages, cancellation, and environment filtering.
- Add an ignored integration test for initialization from a relocated prepared
  runtime and rejection of duplicate initialization.

GitOrigin-RevId: 8fff68fe26e52cb2e0722dc98fb5f124eddb0d4d
2026-09-03 22:53:45 +00:00
felixxia-oai 1d74c3ba1e Persist verified user answers in Guardian thread context (#42579)
## Why

Guardian authorization reviews need host-verified `request_user_input` answers to
remain aligned with thread history across compaction, resume, and rollback.

## What changed

- Record verified question-and-answer pairs in retained thread context when
  `GuardianThreadContext` is enabled, and source both local and delegated Guardian
  review evidence from that context.
- Preserve complete answer records within the evidence budget. Mark authorization
  context incomplete when records are unavailable or oversized, and defer cached
  fast approvals in that case.
- Keep the existing runtime-only answer path for threads that do not enable the
  thread-owned context.

## Testing

- Cover answer retention through compaction and resume, removal after rollback,
  source-call eviction, oversized answers, delegated review evidence, and legacy
  behavior.

GitOrigin-RevId: 01f6f4541515434a0d00fdde318ff59f2f8bd2eb
2026-09-03 16:41:11 +00:00
Eric Traut b7f710273e Support the app-server daemon on Windows (#42405)
## Why

The managed app-server daemon and automatic TUI attachment were limited to
Unix, so Windows users could not share a background server across Codex
sessions.

## What changed

- Enable daemon lifecycle commands, `codex agents` startup, and AF_UNIX daemon
  discovery on Windows.
- Require Windows daemons to start non-elevated and verify the discovered
  socket directory and peer before connecting.
- Preflight detached process support, preserve path-based environment settings,
  and keep daemon processes out of the launching project's working directory.
- Fall back to an embedded server when an implicitly discovered daemon cannot
  connect, while preserving errors for explicit remote endpoints and bypassing
  daemon reuse when `CODEX_EXEC_SERVER_URL` is set.

## Testing

Add coverage for Windows launch restrictions and socket validation, implicit
connection fallback, executor selection, daemon discovery, and the agents
dashboard startup result.

GitOrigin-RevId: e7fd63cf8690bcca8b5155266416f31f2f792aa2
2026-09-03 01:26:35 +00:00
Matthew Zeng a28aab7587 Update rmcp to 3.2.0 (#42383)
## What changed

- Upgrade `rmcp` and `rmcp-macros` from 3.1.3 to 3.2.0.
- Refresh the Cargo and Bazel lockfiles for the new versions.

GitOrigin-RevId: a1d7696eafe6a150c6e09c3cc9c3663a8fd68679
2026-09-02 23:13:33 +00:00
Eric Traut 715294448f Support managed app-server lifecycle on Windows (#42381)
## Why

The PID-managed app-server lifecycle previously supported Unix only, leaving Windows unable to start and stop the managed daemon.

## What changed

- Launch the Windows daemon as a detached process and fail clearly when the host Job Object prevents detachment.
- Add Windows file locking and process creation-time checks so stale or reused PIDs cannot cause an unrelated process to be terminated.
- Request graceful shutdown through a shutdown file, then terminate the verified process after the grace period.
- Prepare private daemon state directories consistently and pin the executable path across installer junction updates.

## Testing

Add Windows tests for process-query permissions, reused and inaccessible PIDs, and stale-record cleanup.

GitOrigin-RevId: 5ea579a7c3dc8bb6122ec3a02513836ccaacc41c
2026-09-02 23:10:08 +00:00
chess 665e5f45ab Clean up Windows sandbox resources on app uninstall (#42375)
## What changed

- Record the authenticated sandbox owner and observe package uninstall events across service restarts and session changes.
- During uninstall, serialize setup and cleanup, disable sandbox accounts, stop their processes, and remove sandbox directories, firewall and WFP rules, hidden-user entries, accounts, and the sandbox group.
- Restrict desktop-owned directory cleanup to pinned paths and perform it while impersonating the authenticated owner.
- Detect missing or disabled sandbox accounts after interrupted cleanup and reprovision them before restoring network access.

GitOrigin-RevId: 7d63fff7ddcff3eb01d018653146df12044bd277
2026-09-02 22:19:51 +00:00
viyatb-oai f53c91be2c Add attributed exec process lifecycle telemetry (#42373)
## Why

Exec-server process and network-policy events need stable attribution to the
launching tool call and executor without recording process payloads.

## What changed

- Add optional `ExecMetadata` to `ExecParams` and propagate the thread and tool
  call IDs from unified exec.
- Emit bounded OpenTelemetry events for process start, spawn failure, sandbox
  denial, and exit. Correlate them with the launch trace and registry-issued
  executor identity while excluding arguments, paths, environment values,
  output, and error text.
- Preserve launch attribution across long-running processes and reconnects, and
  attach the same metadata to network-policy audit events.
- Keep the protocol backward compatible when metadata is omitted, and prevent
  invalid trace headers from inheriting an unrelated active span.

## Testing

Add coverage for metadata serialization and propagation, lifecycle event
fields, trace relationships, reconnect behavior, spawn failures, sandbox
denials, and network-policy attribution.

GitOrigin-RevId: 7aa480a7289c73cb95e2c124c35500bb6f0d5084
2026-09-02 21:58:58 +00:00
johnl-oai c4ea7294b9 Prepare managed policy validation for Windows sandbox provisioning (#42344)
## What changed

- Add a Windows sandbox service validator that loads managed configuration while impersonating the provisioning client.
- Reject elevated sandbox or network settings that conflict with managed requirements, including local binding and HTTP or SOCKS proxy-port restrictions.
- Add a one-shot cloud configuration loader that bypasses the disk cache so policy checks use a fresh backend response without modifying cached configuration.
- Keep the provisioning integration disabled until authenticated transport can supply the policy inputs.

## Testing

- Cover elevated sandbox restrictions, disabled networking, local binding, proxy-port classification, malformed policy, impersonation failure, and cache bypass behavior.

GitOrigin-RevId: b76aa8959c515c6507c2eaf61af768efa1305253
2026-09-02 20:07:11 +00:00
johnl-oai add870a4bf Harden Windows sandbox provisioning file handling (#42342)
## Why

Windows sandbox provisioning performs privileged work beneath a requested Codex home. Provisioning paths and output entries must remain bound to the files and directories that were validated, without following reparse points or writing through caller-controlled hard links.

## What changed

- Add reusable no-reparse directory opens, path validation, directory guards, and handle-relative atomic file replacement.
- Use fresh or retained handles for provisioning logs, sandbox credentials, setup markers, and error reports in service provisioning mode.
- Add service-side validation for framed provisioning requests, proxy settings, fixed-drive Codex homes, and pinned sandbox directories.

## Testing

Add Windows tests covering junction and unsafe-path rejection, hard-link-safe output replacement, retained output handles, request framing and normalization, and ancestor pinning.

GitOrigin-RevId: 2ec4f0d15404c3387df1414bb0c1915cf452ecf2
2026-09-02 20:03:43 +00:00
johnl-oai 501931b399 Add Windows sandbox service lifecycle scaffolding (#42341)
## What changed

- Add the `codex-windows-sandbox-service` workspace crate and Windows-only build target.
- Integrate `CodexSandboxService` with the Windows Service Control Manager, including start, stop, shutdown, status reporting, and event-log handling.
- Support service mode by default and a debug-only `--foreground` mode. Keep provisioning IPC disabled until authenticated request handling is available.

## Testing

- Add argument-parsing tests for service, foreground, unknown, and multiple arguments.

GitOrigin-RevId: 1c84f0a593b65e5cf97ab7606cde03675542b0b0
2026-09-02 19:49:58 +00:00
Eric Traut 73e94ee7a6 Harden Windows control socket rendezvous (#42326)
## Why

Windows control sockets need a rendezvous directory that cannot be accessed or
replaced by another user while the listener is active.

## What changed

- Create socket directories with a protected, inheritable, current-user-only
  DACL, and reject existing directories with broader permissions.
- Validate socket paths without following a junction at the rendezvous
  directory, and pin the directory through listener cleanup to prevent
  replacement after validation.
- Add a Windows peer check that verifies the kernel-reported peer process uses
  the current user's non-elevated token.
- Document the Windows requirements for custom control socket paths.

## Testing

Add Windows coverage for ACL inheritance and rejection, junction and volume-root
rejection, extended-length paths, directory pinning, and elevated peer rejection.

GitOrigin-RevId: cc8dd76e5732c288f6346cfd6138051188533bfa
2026-09-02 18:42:06 +00:00
Felipe Coury 637c3227b3 Avoid executing PATH helpers before workspace trust (#42324)
## Why

Automatic startup work and `codex doctor` can run before a workspace is
trusted. A repository-controlled `PATH` must not be able to make those flows
execute workspace-provided helpers.

## What changed

- Resolve helpers used by automatic startup from trusted system installation
  directories, and give plugin-sync Git subprocesses a sanitized environment.
- Make terminal detection environment-only and have doctor inspect executable
  locations without running them. Fetch update metadata with the HTTP client
  instead of `curl`.
- Fall back conservatively when trusted terminal helpers are unavailable,
  including for tmux keyboard enhancement flags.

## Testing

Add black-box coverage with hostile workspace `PATH` entries for startup,
interactive tmux startup, support log collection, doctor, and curated plugin
sync. Add unit coverage for trusted executable resolution and bounded update
HTTP responses.

GitOrigin-RevId: 3b8995eb422b60ed53b0386951de59e8f9bfc542
2026-09-02 18:34:27 +00:00
felixxia-oai 5971d42847 Preserve verified answers across history compaction (#42293)
## Why

Compaction replaces the model history window, but host-verified user answers may
still be needed for the rest of the owning thread.

## What changed

- Add bounded, model-invisible retained context for verified
  `request_user_input` answers.
- Persist retained context in rollout events and compaction checkpoints, and
  restore it when a thread resumes.
- Remove answers when their source turns are rolled back, and prevent
  parent-local answers from being inherited by child forks.

## Testing

Add lifecycle coverage for deduplication and size limits, compaction and resume,
rollback, rollout migration, and child forks.

GitOrigin-RevId: c32900d3e8120a9899b71f0576683acd67a4b898
2026-09-02 15:07:11 +00:00
felixxia-oai 8d32abcd01 Report the exec-server release version in environment info (#42270)
## Why

Clients need the executor's release version to make version-based compatibility decisions.

## What changed

- Add `executorVersion` to `EnvironmentInfo` and return it from both `initialize` and `environment/info`.
- Resolve the package release version at process startup so the advertised value remains stable for the server's lifetime.
- Use `0.0.0` when the version is unknown or omitted by a legacy executor.

## Testing

- Cover packaged and source-build initialization, startup caching, protocol serialization, and client-side metadata caching.

GitOrigin-RevId: a0c7f0b701a8e7e50345cd5fb1b5c5d2d38ff3b1
2026-09-02 12:07:32 +00:00
pmccrary-oai bdfd769640 Track history notes thread hint outcomes (#42247)
## What changed

- Emit a `codex_thread_hint_status` analytics event for each native history-notes thread hint attempt.
- Report whether retrieval was successful or failed along with thread context and timing, without including hint contents.
- Treat valid empty responses as successful retrievals while continuing to omit them from the context window.

## Testing

- Extend the app-server history-notes tests to verify success, empty-result success, and backend failure statuses.

GitOrigin-RevId: b9462f312847e8c871ed2be0c8cf8df0928a7fdd
2026-09-02 09:12:17 +00:00
Benjamin Carlsson 798833fe97 Add managed worktree creation (#42196)
## What changed

- Add `WorktreeManager::create` to create detached, Desktop-compatible worktrees from `HEAD` or an explicit base while preserving the source working-directory path.
- Isolate worktree Git operations from inherited repository selectors, hooks, filesystem monitors, and configured content filters.
- Validate the destination working directory and roll back incomplete worktrees and empty allocation buckets on failure.

## Testing

- Cover layout, base selection, annotated tags, nested working directories, Git environment isolation, filter suppression, source checkout preservation, unsafe symlinks, and creation rollback.

GitOrigin-RevId: bf172c3ff4268dab603d00a1d547485fcd0de368
2026-09-02 03:46:24 +00:00
willwang-openai 6b59cefcbb Support remote marketplaces in the plugin CLI (#42150)
## What changed

- Include remote catalog entries in `codex plugin list`, including their source,
  version, install policy, and authentication policy in JSON output.
- Support adding and removing remote plugins through the existing plugin CLI.
- Cache remote catalogs by scope and collection. Prefer fresh cached results, and
  refetch once when an add request misses a plugin in the cache.
- Preserve the local curated catalog when an unfiltered remote listing fails,
  while surfacing errors for explicitly selected remote marketplaces.

## Testing

- Cover remote listing, installation, removal, catalog fallback, cache refresh,
  collection isolation, and install failure behavior.

GitOrigin-RevId: 09796b2c393d102e00ba9289f784d78a2e166a54
2026-09-01 22:51:14 +00:00
jif 86b7d12742 Add Guardian V2 analytics events (#42144)
## What changed

- Emit `codex_guardian_v2_classification` events with the outcome, risk
  level, duration, model, and thread attribution.
- Emit `codex_guardian_v2_fast_decision` events when Guardian V2 approves an
  action without a full review.
- Add `guardian_v2_enabled` to turn analytics, accounting for review scope
  and model requirements.

## Testing

- Extend the Guardian V2 app-server coverage to verify event payloads,
  attribution, enablement state, and scoped event emission.

GitOrigin-RevId: 0805975f8e8f9d954aa1058ea8c60545a090ae74
2026-09-01 22:02:42 +00:00
Charlie Marsh 8436b749a4 Bound Git root discovery for metadata enrichment (#42132)
## Why

Git root discovery is optional metadata work, but filesystem probes can block. They should not exhaust Tokio's blocking pool, delay runtime shutdown, or prevent later turns from observing repository changes.

## What changed

- Add a shared `GitRootDiscovery` service that coalesces concurrent lookups for the same working directory and limits probes across directories.
- Run probes on detached threads, retain in-flight work across caller cancellation, and discard completed results instead of caching them.
- Use the service for turn and memory metadata enrichment, abort unused turn enrichment when its state is dropped, and limit memory metadata waits to one second.

## Testing

Add coverage for probe sharing, capacity limits, cancellation, fresh discovery, runtime shutdown, memory timeouts, and repositories restored after startup prewarming.

GitOrigin-RevId: bca46fc263e7a12a2f69146d8a0b3e7c7e0846cb
2026-09-01 20:41:32 +00:00
Adam Perry @ OpenAI 9969043b95 Extract OTEL trace WebSocket into a reusable crate (#42102)
## What changed

- Add `codex-otel-trace-websocket` with a `TraceWebSocket` API that owns the
  loopback OTLP receiver and WebSocket listener.
- Bind both listeners during startup, expose the exporter and bound listener
  addresses, and surface listener failures through a single lifecycle method.
- Update `codex-code-mode-host` to use the new crate and shut down the bridge
  after flushing its trace provider.

GitOrigin-RevId: ec5ca4c4369b6b3b1232875c2699730eb748ab1c
2026-09-01 17:17:45 +00:00
iceweasel-oai ade0ccacf9 Record Windows MXC availability (#42094)
## What changed

- Probe whether the native Windows MXC process security environment is usable
  when selecting a sandbox.
- Emit `codex.windows_mxc.available` at most once per process, labeled with the
  probe result.
- Add the MXC workspace dependencies for Cargo and Bazel, and pin
  `tracelogging` to a version compatible with the GNU Windows toolchain.

GitOrigin-RevId: d7cfeed6f0333feda5764e1ee44fe5bbd50f9a58
2026-09-01 16:40:05 +00:00
jif 9d0eae74cd Include shared histories in rollout compression (#42039)
## What changed

- Make `local_thread_store_compression` compress cold rollout files across shared and forked histories without a separate compression mode.
- Retire `local_thread_store_shared_compression` while continuing to accept it in strict configuration without changing compression behavior.
- Read rollout files through the compressed-rollout reader when `codex exec resume` determines the latest turn's working directory.

## Testing

- Cover compression and restoration of an archived fork chain.
- Cover cwd-based resume selection when only the compressed rollout remains.
- Cover the removed compatibility key in strict configuration.

GitOrigin-RevId: 8edebf61eccc1f79cabb8e641f8982c76f337bf1
2026-09-01 12:23:20 +00:00
Adam Perry @ OpenAI 55e5158e18 Improve tracing for nested tool calls and exec processes (#41950)
## Why

Code-mode callbacks can outlive their initial request and run in separate tasks,
so their trace context must be preserved explicitly.

## What changed

- Preserve the execution context for code-mode callbacks and add spans for
  nested tool invocations in the runtime and dispatch broker.
- Propagate each tool invocation span as the parent of its streamed gRPC
  callback, falling back to the outer execution context when needed.
- Add a `codex.exec_server.process` span for each process lifetime, including
  its process ID and completion result, without retaining the request span.

GitOrigin-RevId: a6059e34d895416f5517e51dad6ca0078355adca
2026-09-01 01:00:50 +00:00
Jennifer Zhao 981d6b652b Omit undersized WAV output from Code Mode (#41934)
## Why

Audio models cannot reliably encode tool-generated clips shorter than 25 ms.

## What changed

- Measure the available audio frames in base64-encoded PCM and IEEE float WAV outputs.
- Replace clips under 25 ms with explanatory text while preserving surrounding Code Mode output.
- Leave clips at or above the threshold and unrecognized audio formats unchanged.

## Testing

- Cover the duration boundary across multiple sample rates and all supported `audio()` input forms.
- Verify bounded and truncated WAV chunks, output after `yield_control()`, and end-to-end mixed text and audio output.

GitOrigin-RevId: 342b4bdbc65db4238580f9cdc372763c2576eca5
2026-08-31 23:36:04 +00:00
Tamir Duberstein 9a4b78579a Move the config schema generator into a dedicated crate (#41915)
## What changed

- Add the `codex-config-schema` workspace and Bazel crate for the
  `codex-write-config-schema` binary.
- Preserve `codex-rs/core/config.schema.json` as the default output location.
- Update `just write-config-schema` to run the new crate and remove the schema
  generator's `clap` dependency and binary target from `codex-core`.

GitOrigin-RevId: 7447f97d6e44d1077b16d2d850f8a8add9117ea6
2026-08-31 21:27:51 +00:00
Benjamin Carlsson 13bc770eaf Add installed voice host lifecycle support (#41902)
## What changed

- Add `VoiceHost` to resolve the packaged voice helper, launch it with an
  allowlisted environment, perform the protocol handshake, and enforce bounded
  shutdown and process cleanup.
- Preserve native executable path encoding in the pipe process APIs.
- Add `third_party/voice/assemble_package.py` to create a fresh package copy
  containing a target-compatible helper and a provenance manifest with file
  hashes.

## Testing

- Cover installed helper lifecycle, build matching, missing and symlinked
  helpers, non-UTF-8 package paths, environment filtering, package validation,
  target pairing, and failure cleanup.

GitOrigin-RevId: f893074b36ae6bb9bcedc00fbe7af6bb72745f4c
2026-08-31 19:44:44 +00:00
Evan Fannin 53691bad98 Keep MCP event subscriptions alive after task unloading (#41899)
## Why

Event subscriptions may need to outlive the MCP runtime associated with a task.

## What changed

- Add `McpEventStreamOpener`, which retains the hosted Apps MCP connection settings and opens each event stream with an independent client.
- Keep subscription cancellation tied to account changes and removal of the task's event server, including when cancellation ownership is forwarded across runtime replacement.
- Expose the opener and cancellation forwarding through `McpResourceClient`.

## Testing

Add an HTTP transport test that shuts down the runtime, refreshes credentials for the same account, and verifies that a new event stream connects with the current authorization and receives notifications.

GitOrigin-RevId: e102aa382e63121661ff0567f7523515e459922d
2026-08-31 19:18:18 +00:00
Benjamin Carlsson d60560f14e Add the voice helper lifecycle foundation (#41897)
## What changed

- Add `codex-realtime-webrtc` with a length-prefixed, 256-byte JSON control protocol for `hello`, `ready`, `close`, and `closed` messages.
- Add a hardened `codex-voice-host` binary that validates protocol and build compatibility, handles orderly shutdown and parent pipe loss, and exposes its build commit.
- Keep this initial foundation limited to process lifecycle management; it does not open audio devices, load native plugins, negotiate WebRTC, or enable voice in the TUI.

## Testing

- Exercise the helper as a subprocess, covering handshake, explicit close, parent pipe loss, malformed frames, incompatible builds, and invalid message order.

GitOrigin-RevId: 6d8d7dd7ba5c4886879882d79314d235f6ab7cf4
2026-08-31 19:08:43 +00:00
felixxia-oai 9f97cb79eb Preserve Guardian review evidence across compaction (#41879)
## Why

Guardian reviews need the original conversation evidence even after the parent context has been compacted, while rollback must still remove evidence from reverted turns.

## What changed

- Build synchronous and asynchronous Guardian transcripts from the retained review-history snapshot.
- Use the shared Guardian transcript collector for consistent entry formatting, tool attribution, and per-entry truncation while preserving the existing transcript retention budgets.

## Testing

- Add app-server coverage for retaining user and tool evidence after compaction and discarding it after rollback.
- Extend transcript tests for standalone tool outputs, shell calls, attribution, truncation, and retention limits.

GitOrigin-RevId: a18b1176f14a4d006c263aa82d4e593afee66bc3
2026-08-31 17:10:38 +00:00
jif 1c1e17782a Preserve Guardian review evidence across compaction (#41846)
## Why

Context compaction can replace original conversation items that Guardian still needs to review approval-sensitive actions.

## What changed

- Retain a bounded, chronological review history independently of the model's compacted history.
- Keep separate limits for user messages and other transcript items so tool traffic cannot evict user instructions.
- Use the retained evidence for synchronous and asynchronous Guardian transcripts, while resetting it after rollback or history reconstruction.

## Testing

- Cover retention through compaction and eviction, reset behavior after rollback, independent size limits, and asynchronous transcript construction.

GitOrigin-RevId: c5851e992820504fc18de8cfec7f6f9f1b69f316
2026-08-31 13:57:59 +00:00
Tamir Duberstein e45226e771 Use the async stack budget for approval reviews (#41840)
## What changed

- Define the shared 16 MiB thread stack budget in `codex-async-utils`.
- Use it for both the async main/runtime threads and the dedicated approval-review thread.

GitOrigin-RevId: c825669e1545bcfaeaaaa07066dc63874f46630f
2026-08-31 13:44:12 +00:00
Eric Traut 94cbbddafc Support package-style MCP server names (#41700)
## What changed

- Allow MCP server names to contain `:`, `@`, `/`, and `.`, enabling names such as `npm:@modelcontextprotocol/server-sequential.thinking`.
- Preserve these names across `mcp add`, `get`, `list`, and `remove`, runtime tool namespaces, and OAuth credential lookup.
- Quote non-bare server names in generated `config.toml` recovery hints and keep similarly escaped OAuth credential names isolated.

## Testing

- Add CLI and runtime round-trip coverage for npm-style names.
- Add snapshot coverage for quoted recovery hints and a regression test for OAuth credential-name collisions.

GitOrigin-RevId: 97e9353cdbc544db4eb2377ddbfa1d2d539b1e04
2026-08-30 16:45:34 +00:00
chess aaa7ed042e Harden diagnostic report uploads (#41569)
## What changed

- Send the core report event before its attachments, then upload each attachment in a separate gzip-compressed envelope linked to that event.
- Bound encoded and decoded payload sizes, truncate oversized attachment copies with format-aware handling for JSONL, and preserve attachment priority when applying size limits.
- Retry transient attachment failures within a shared upload-time budget while honoring Sentry rate-limit and retry headers without replaying the core event.
- Add APIs to prepare and send persisted report events and attachments using a stable report ID, explicit delivery outcomes, and caller-controlled retries.

## Testing

- Cover attachment retries, truncation, rate limits, rejected responses, transport failures, redirect handling, content types, diagnostic consent, and report tags.

GitOrigin-RevId: 5eacd384843795d05d1c024b76e1c6d0c1c4b8d7
2026-08-29 20:26:42 +00:00
Samuel Yuan 2181224dad Support app targets in executor plugin hooks (#41456)
## What changed

- Admit the curated remote Browser plugin's `Stop` and `SubagentStop` hooks when its listed `browser.turn_ended` tool matches the expected connector and is enabled by app policy.
- Carry trusted app routing metadata and the MCP environment through hook registration and execution.
- Register hooks from multiple executor environments, while deduplicating cleanup calls that share an event, MCP environment, server, and tool.

## Testing

- Cover Browser and Computer Use cleanup through separate MCP routes, including user- and managed-policy disablement.
- Verify routing metadata, environment selection, multi-environment registration, and target deduplication.

GitOrigin-RevId: 66de088fd201b9ed0db8b97c23d33de88135120d
2026-08-29 03:14:35 +00:00
felixxia-oai 0918cd2c08 Add shared Guardian transcript collection (#41422)
## What changed

- Add a reusable transcript contributor for synchronous and asynchronous Guardian context, with borrowed conversation history and request-specific configuration.
- Preserve conversation order and role or tool attribution while applying separate per-entry limits for messages, ordinary tools, and Node REPL-backed outputs.
- Make tool calls, tool outputs, and reasoning independently configurable, and expose transcript collection without requiring section composition.
- Register the transcript contributor in a process-wide default registry that retains no request state.

## Testing

- Add unit coverage for role and tool attribution, optional evidence sources, orphaned and named tool outputs, registry reuse, and request-specific entry limits.

GitOrigin-RevId: 2b60f8db4618de8a7175f8ee0183375a526a87db
2026-08-28 21:43:21 +00:00
felixxia-oai 4878401e8f Add shared Guardian context primitives (#41392)
## What changed

- Add the `codex-guardian-context` crate for assembling structured context shared by synchronous Guardian review and asynchronous scoring.
- Preserve transcript entry roles and original byte counts, and let section contributors declare shared or consumer-specific scope.
- Collect applicable sections in registration order, skip optional sections, and fail the collection when required evidence is missing.
- Provide UTF-8-safe prefix/suffix truncation with approximate omitted-token accounting.

## Testing

- Cover scoped registry collection, ordering, optional sections, required-evidence failures, truncation markers, and UTF-8 boundaries.

GitOrigin-RevId: 19d8458403c470c9b413992dbff925a12259c3c5
2026-08-28 19:02:26 +00:00
Eddie Chen 60fc699560 Give Guardian classifications distinct turn identities (#41385)
## What changed

- Generate a UUIDv7 turn ID for each Guardian classification and preserve it across retries.
- Record the owning turn as `parent_turn_id` and propagate its trusted `root_turn_id` when known, including through the tool lifecycle extension API.
- Omit ambiguous or unknown root lineage and keep classification identities distinct when reusing Guardian sessions and WebSocket connections.

## Testing

- Cover classification metadata across retries, connection reuse, concurrent samples, auth refresh, and consecutive owning turns.

GitOrigin-RevId: 78757f2ad031844e4b545802f634c678a9ee2298
2026-08-28 18:23:47 +00:00
Brandon Zhang 94311d4475 Forward history note images to the model (#41292)
## What changed

- Convert history backend `images` into `input_image` function-call output items alongside encrypted or plaintext history content.
- Keep image data out of logged output and post-tool-use hook responses.
- Reject malformed image attachments instead of silently dropping them.

## Testing

- Cover encrypted and plaintext outputs, supported image detail values, malformed attachments, and propagation into the next model request.

GitOrigin-RevId: a8eee10eb09c637bcfb06a759dd6313caf707e51
2026-08-28 07:18:54 +00:00
yqz-openai eed1dee69e Propagate trace context through gRPC code mode (#41017)
## What changed

- Inject W3C `traceparent` metadata into code-mode session and execution requests.
- Carry each execution's trace context on streamed tool callbacks so callback and nested-tool spans remain connected across the gRPC boundary.
- Preserve the relevant spans when execution, callback, and dispatch work moves into spawned tasks.

GitOrigin-RevId: 6482963cb9b2e2f0ea27f1345bba208b70a80d0b
2026-08-27 01:55:53 +00:00
felixxia-oai b68acc4d4b Trust invoked user skills in Guardian reviews (#41006)
## Why

Guardian treated all skill instructions as untrusted, so it could not use the
invocation of a user-owned skill as authorization evidence.

## What changed

- Record explicit and implicit skill invocations through extension contributors.
- Send Guardian a bounded, deduplicated list of canonical skill paths only when
  they resolve under the user's Codex or `.agents` skill directories.
- Keep repository skills and symlinks that escape trusted roots untrusted, and
  send only verified paths rather than skill contents to the classifier.
- Avoid recording host skill prompts that were superseded by core skill
  injection.

## Testing

- Added unit coverage for trusted roots, symlink escapes, deduplication, and
  size limits.
- Added integration coverage showing that Guardian trusts invoked user skills
  while rejecting a forged repository skill.

GitOrigin-RevId: 1cc0be230f7e6649110201c23c3de8f26447ff36
2026-08-27 00:07:26 +00:00
andrewgu-oai f3741880f6 Harden managed proxy listener handoff (#40999)
## Why

Linux managed networking routed proxy connections through filesystem-backed Unix sockets, requiring an extra readable sandbox path and stale socket cleanup.

## What changed

- Transfer loopback TCP listeners from the sandbox namespace to host bridge processes over anonymous Unix socket pairs.
- Validate transferred descriptors, keep bootstrap channels private to each sandbox launch, and close them before running untrusted commands.
- Tie proxy bridge lifetime to the sandbox so cancellation closes active upstream connections.

## Testing

- Cover descriptor validation and interrupted handoffs.
- Verify concurrent sandboxes keep proxy endpoints isolated and do not inherit privileged sockets.
- Exercise managed networking end to end through `app-server`, confirming proxied access succeeds while direct host access remains blocked.

GitOrigin-RevId: 6561e09c83b927c604951bf2aa00ff532d179da5
2026-08-26 23:41:18 +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
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