Commit Graph
4198 Commits
Author SHA1 Message Date
jif 6c729ef1c1 Refresh MCP runtimes when session auth changes (#35146)
## What changed

- Track the authentication state used by each published MCP runtime and mark the runtime dirty when the auth mode, account details, or token changes.
- Refresh plugin auth mode before evaluating pending MCP work, and rebuild the initial MCP projection if authentication changes during session startup.
- Carry one auth snapshot through runtime construction so plugin discovery and MCP connections use consistent credentials.

## Testing

- Add a regression test that replaces an API key, reloads authentication, and verifies that MCP refresh publishes a runtime using the new credentials.

GitOrigin-RevId: 219c2f7a157f53b5467116e18bfe70a8c3dca695
2026-07-24 09:50:41 +00:00
jif ef2d3edb95 Prewarm MCP runtime updates in the background (#35144)
## Why

Changes to MCP inputs can require rebuilding the runtime before the next model
step. Start that work eagerly so the step is less likely to wait for it.

## What changed

- Add a per-session worker that prewarms dirty MCP state after configuration,
  authentication, or explicit refresh changes.
- Coalesce queued refreshes and always rebuild from the latest session state,
  while keeping exact model-step refreshes as the correctness path.
- Stop and join the worker during session shutdown.

## Testing

- Verify rapid refreshes skip superseded configuration and publish the latest
  state.
- Verify MCP-relevant policy changes mark the runtime for refresh.

GitOrigin-RevId: 045ed406caff97dcbb483bc183fd16eae0c7d7ad
2026-07-24 09:40:10 +00:00
TAFOYA-OAI 81da9deb06 Allow hosts to customize wait_for_environment descriptions (#35106)
## What changed

- Add `WaitForEnvironmentToolConfig` as thread extension data for overriding the
  model-visible tool and `environment_id` descriptions.
- Preserve the default descriptions when no override is provided or when the
  configured descriptions exceed the input or serialized tool-spec limits.
- Keep `wait_for_environment` availability gated by the deferred executor
  feature independently of whether a host override is present.

## Testing

- Cover default, custom, oversized, and feature-disabled tool configurations.
- Verify the custom descriptions in the deferred-environment integration flow.

GitOrigin-RevId: 6b49a73a434becd99ea5df911be53f3706a17c0a
2026-07-24 06:03:36 +00:00
Channing Conger f61b51ddd9 Support remote code-mode hosts in app-server (#35098)
## What changed

- Add `--code-mode-host ws://...` and `wss://...` support to `codex app-server`, gated by the `code_mode_host` feature. When omitted, app-server continues to start a local host.
- Share one remote WebSocket connection across the process's threads, using the configured HTTP client's proxy and TLS policy and preserving the existing framed host protocol.
- Reject invalid host URLs, bound WebSocket frame sizes, close connections cleanly, and return an error when a connection exceeds 1,024 pending delegate calls without disconnecting it.

## Testing

- Cover CLI validation, WebSocket protocol execution and shutdown, connection sharing across app-server threads, and delegate-call capacity recovery.

GitOrigin-RevId: 715e82d4d9db1e7e2f91b754a777dcab504e2ae4
2026-07-24 04:37:01 +00:00
Ahmed Ibrahim 3947f0d0c3 Avoid duplicating deferred sources in tool search (#35065)
## Why

Deferred tool world state already advertises the available tool sources, so
repeating them in the `tool_search` description adds redundant context.

## What changed

- Omit the source listing from `tool_search` when
  `DeferredToolWorldState` is enabled while preserving the tool discovery
  instructions.
- Include the source-listing mode in the tool search handler cache key so the
  description is rebuilt when the feature state changes.

## Testing

Added coverage for source omission, cache invalidation when the feature is
toggled, and the resulting request payload.

GitOrigin-RevId: f7dc0c4f3351a7c7daadec936b26a080202730bc
2026-07-24 01:23:49 +00:00
Ahmed Ibrahim 1d4b58f32d Track deferred tool namespaces in world state (#35063)
## What changed

- Add the disabled-by-default `deferred_tool_world_state` feature, which exposes deferred tool namespaces and their descriptions to the model in a `<tools>` world-state section.
- Emit added and removed namespace updates as tool availability changes, bound the rendered context size, and persist nonempty state across thread resumes.
- Omit empty tool state while retaining deferred tool discovery through `tool_search`.

## Testing

- Cover initial, unchanged, removed, recovered, empty, and resumed namespace state, along with description truncation and rendered-size limits.

GitOrigin-RevId: 867e599666dc3569eb0789ba78aaa40254253a6f
2026-07-24 01:20:04 +00:00
tongzhou wang fb4e6ba2f4 Allow disabling the update_plan tool (#35054)
## What changed

- Add a default-on `tools.update_plan.enabled` configuration option.
- Omit `update_plan` from the visible and registered tool sets when the option is disabled.

## Testing

- Cover configuration resolution and tool registration for the disabled setting.

GitOrigin-RevId: c13aa463a6911956fca9f0ef5b74841b543798c4
2026-07-24 00:10:22 +00:00
Won Park bb24b67d33 Register the Guardian V2 feature flag (#35049)
## What changed

- Add `GuardianV2` to the feature registry for automatic approval reviews.
- Expose it as `features.guardianv2` in the configuration schema.
- Keep the under-development feature disabled by default.

GitOrigin-RevId: 92fa107e3ff1396a75a28a071353743a7bb48c43
2026-07-24 00:00:43 +00:00
iceweasel-oai b115de97d7 Preserve Windows sandbox proxy settings in guardian sessions (#35036)
## Why

Guardian review commands can run without the proxy-port environment used to
configure the parent session. Reconciling persistent Windows sandbox settings
for those commands can discard the parent's proxy configuration.

## What changed

- Add a session-level Windows sandbox proxy-settings mode and use `Preserve`
  for guardian review sessions while keeping `Reconcile` as the default.
- Carry the mode through unified exec and the exec-server sandbox context to
  Windows process launches.
- In preserve mode, reuse the proxy settings recorded by the existing sandbox
  setup marker.

## Testing

Added coverage for guardian session configuration, exec-server transport, and
preserving an existing setup marker when proxy-port environment variables are
absent.

GitOrigin-RevId: 214655c6e6d97360906431773dc0de0fdda2db6e
2026-07-23 23:35:29 +00:00
rafael-oai 41775559ca Expose Browser Use requirements through the app server (#35033)
## What changed

- Parse the `browser_use.disable_auto_review` setting from layered
  `requirements.toml` configuration.
- Return the setting as `browserUse.disableAutoReview` from
  `configRequirements/read` and publish it in the generated JSON and TypeScript
  schemas.

## Testing

- Add an app-server RPC test covering the Browser Use requirement.

GitOrigin-RevId: 5749d5bc17bcc5b582bf7ed59b8e5b72d6c8f7fc
2026-07-23 23:21:33 +00:00
Kyle Brown 9fc4e5a7aa Preserve plugin attribution across command approvals (#35029)
## What changed

- Add optional `plugin_id` and `script_path` fields to execution approval and guardian assessment events.
- Propagate validated plugin attribution through delegated approvals, guardian-reviewed command items, app-server notifications, thread history, and rollout traces.
- Preserve attribution on both started and completed command items, including declined commands.

## Testing

- Extend core, app-server, thread-history, and rollout-trace tests to cover plugin attribution propagation and serialization.

GitOrigin-RevId: 723684d010cab04142918d7a95e06ed95d008da1
2026-07-23 23:13:59 +00:00
jif 091e4a5d7c Preserve refreshed Apps tools across MCP runtime updates (#35028)
## Why

After a remote plugin install refreshes the Apps tool catalog, a later MCP
runtime publication must not restore the catalog from the previous connection.

## What changed

- Rebuild the MCP runtime with fresh connections when hard-refreshing Apps
  tools, using the latest desired runtime configuration.
- Refresh the Apps catalog on the newly published connection so subsequent
  runtime updates retain it.

## Testing

Extend the remote plugin install test to verify that both newly available and
missing Apps tools remain consistent after an unrelated runtime configuration
refresh.

GitOrigin-RevId: 5b675d53d56379ab67fab09512fc2ad0ffcb8535
2026-07-23 23:10:00 +00:00
Kyle Brown 84fa68b429 Attribute command executions to trusted plugin scripts (#35020)
## What changed

- Resolve shell and unified-exec commands against the trusted plugin roots loaded for each turn.
- Add optional `pluginId` and safe plugin-relative `scriptPath` fields to command execution items and legacy execution events, and propagate them through app-server notifications.
- Include the attribution in command execution analytics while rejecting absolute, unsafe, and unattributed script paths.

## Testing

- Cover attribution for cached curated and remote plugin scripts from command execution through core and app-server events.
- Verify analytics serialization and unsafe-path filtering.

GitOrigin-RevId: 02fac3a233284ccfc6642fa502a95f1881dba83d
2026-07-23 22:31:53 +00:00
Eric Traut 7d4b417cd1 Keep side conversations open when switching threads (#35011)
## What changed

- Add a configurable `toggle_side_conversation` TUI action, bound to `ctrl-/` by default, to switch between a side conversation and its parent without closing either.
- Show the active binding in the footer and distinguish switching from closing with `ctrl-c`.
- Allow starting a replacement side conversation from the parent, while cleaning up the previous side thread and preserving it if cleanup fails.
- Clean up retained side conversations when shutting down the current thread.

## Testing

- Cover shortcut remapping, conflicts, and compatibility with existing `ctrl-/` and `ctrl-7` bindings.
- Cover side-thread replacement and footer states for both the parent and side conversation.

GitOrigin-RevId: a621c21e1c45ba73197d59750b2e05efa5ee1081
2026-07-23 21:03:40 +00:00
felixxia-oai 2c92af09cf Warn when skill catalogs exceed their context budget (#34997)
## Why

Skill catalog rendering can shorten descriptions or omit enabled skills to fit
the model context budget. Surface these reductions so users know when the
model-visible catalog has lost detail.

## What changed

- Emit a warning with the omitted skill count whenever entries do not fit.
- Warn when description shortening averages more than 100 characters per skill.
- Deduplicate executor catalog warnings across repeated world-state builds in a
  turn.

## Testing

- Cover the description-shortening threshold and omission warning text.
- Verify warnings through extension and production-turn catalogs, including
  per-turn deduplication.

GitOrigin-RevId: bd7fc3482e5dfd8c79072f772a2f54aa502478d9
2026-07-23 19:44:12 +00:00
Adam Perry @ OpenAI 7bafdada8b Separate Codex error details from retry metadata (#34996)
## What changed

- Wrap `CodexErrorDetails` and an optional retry delay in `CodexErr`, allowing any mapped error to preserve server-provided retry timing.
- Generate the payload-free `CodexErrKind` classification alongside the error details and reuse it for analytics.
- Update error handling sites to inspect `CodexErr::details()` while preserving existing display, debug, protocol mapping, and retryability behavior.

## Testing

- Add coverage for legacy debug formatting, error-specific retryability, and retry-delay propagation through API error mapping.

GitOrigin-RevId: d3ab8a305f2a2ee21d0c0a8c8c388b06dda9c59a
2026-07-23 19:32:26 +00:00
Celia Chen 265cd2e100 Initialize execution environments with the final HTTP policy (#34995)
## Why

The TUI must inspect the default execution environment before loading its final
configuration. Initializing the environment manager at that point can give
startup services the bootstrap HTTP policy instead of the effective policy after
managed requirements are applied.

## What changed

- Split environment discovery from manager construction so callers can inspect
  the default environment without starting remote connections.
- Build the environment manager after final configuration loading and pass its
  resolved `HttpClientFactory` through all construction paths.
- Add shared test support for managers that use the legacy default HTTP policy.

## Testing

- Cover connection-free environment discovery and explicit HTTP policy
  propagation.
- Verify TUI startup services use the final managed `respect_system_proxy` value.

GitOrigin-RevId: 928fa31e6b4bcfbe1a121cade2f351427fdfa0f4
2026-07-23 19:24:41 +00:00
Adam Perry @ OpenAI c769a05340 Honor the configured SQLite home across state consumers (#34994)
## Why

Codex and SQLite data can use separate home directories, but state consumers
could reconstruct database paths from the Codex home instead of consistently
using the resolved SQLite configuration.

## What changed

- Pass `SqliteConfig` through the core, rollout, state runtime, and thread store
  instead of passing a directory and rebuilding the configuration downstream.
- Use that shared configuration for state, logs, memories, goals, and paginated
  thread-history database access, including integrity checks and cleanup.
- Reject state database handles whose SQLite configuration does not match the
  requesting store.

## Testing

Add coverage with separate Codex and SQLite homes that verifies startup
backfill, thread listing, and paginated history all use the configured SQLite
directory.

GitOrigin-RevId: 1de1cdd1d6ff1d70bbb6c360c8352e6543fb8ebf
2026-07-23 19:19:35 +00:00
sayan-oai 74e9d7efc4 Allow omitting MCP tool prefixes per server (#34991)
## What changed

- Accept a table form of `features.non_prefixed_mcp_tool_names` with an
  optional `server_names` list while preserving the existing boolean form.
- Omit the legacy `mcp__` namespace prefix only for tools from selected MCP
  servers. When no server list is provided, the enabled feature continues to
  omit the prefix for every server.
- Cover configuration resolution, tool normalization, and an MCP stdio
  round trip with selected servers.

GitOrigin-RevId: bdfb7ac54226de5051f06610e2c6b78b23912ef0
2026-07-23 18:54:55 +00:00
Owen Lin 5c94796dc9 Enforce single-writer ownership for paginated threads (#34986)
## Why

Only one app-server process should keep a paginated thread open for writing at
a time, while other processes must still be able to read it.

## What changed

- Acquire a per-thread filesystem lock when creating or resuming a paginated
  thread, and retain it for the lifetime of the live recorder.
- Reject a competing `thread/resume` request with JSON-RPC error `-32600` and
  release ownership when the writer is discarded, deleted, or shut down.
- Remove stale lock files without disturbing active writers.

## Testing

Added coverage for competing processes, ownership transfer after shutdown,
stale-lock cleanup, and lock release during discard and deletion.

GitOrigin-RevId: 81d49a2fffc73ba8135752b4944f27feafa4c0d6
2026-07-23 18:22:41 +00:00
rka-oai 205d37a20f Keep the sleep tool outside code mode (#34969)
## What changed

- Mark `clock.sleep` as `DirectModelOnly` so it remains directly callable in
  code-mode-only sessions without appearing in the nested code-mode tool surface.
- Cover both code mode configurations with a tool-plan test that verifies the
  sleep tool's exposure and visibility.

GitOrigin-RevId: 7ef67bea0678f27fdb17d88004368e5f1d475125
2026-07-23 16:20:30 +00:00
jif e19e65317a Reuse MCP connections across runtime refreshes (#34952)
## Why

Refreshing MCP runtime state should not restart an unchanged, ready server or
relist its tools.

## What changed

- Reconcile refreshed MCP configuration against the published connection set
  and reuse connections whose transport, environment, authentication, OAuth
  credentials, and client capabilities are unchanged.
- Keep tool filters, timeouts, metadata, and plugin provenance in the published
  server view so those settings can change without reconnecting.
- Update elicitation authority in place for reused connections, and reconnect
  when connection-defining inputs or live OAuth credentials change.

## Testing

Added coverage for unchanged-server reuse, view-only and elicitation-policy
updates, connection identity changes, OAuth credential changes, and avoiding
redundant tool listing.

GitOrigin-RevId: f04f5db5fbdde127b6a14f6aa9673112c0b557e7
2026-07-23 14:55:31 +00:00
jif 808d3c2702 Keep session defaults static during config batch writes (#34940)
## What changed

- Treat batch writes containing only `model`, `model_reasoning_effort`,
  `plan_mode_reasoning_effort`, `service_tier`, or `personality` as defaults for
  future sessions, even when `reloadUserConfig` is requested.
- Keep runtime config reloads and skill/plugin cache invalidation for other
  config changes, and document which settings remain session-static.
- Preserve cached skill listings after session-default writes until a forced
  reload, with test coverage for each supported default.

GitOrigin-RevId: 6d988014cf4758ceb9422a11f7c2bd5fc6b9f90a
2026-07-23 13:11:16 +00:00
felixxia-oai 48ebbf5334 Use the API plugin marketplace for Amazon Bedrock (#34931)
## Why

Amazon Bedrock configurations can operate without Codex auth, so auth mode alone
cannot determine which curated plugin marketplace to expose.

## What changed

- Pass the resolved model provider into plugin configuration.
- Select `openai-api-curated` when the resolved provider is `amazon-bedrock`,
  while retaining API-key auth as a selection signal.
- Cover provider overrides and app-server plugin listing without an `auth.json`
  file.

GitOrigin-RevId: 972a22e19c41490d8b19d989da052302dc30e8f2
2026-07-23 11:32:47 +00:00
jif e497325a6a Centralize thread MCP state in McpRuntime (#34930)
## What changed

- Make `McpRuntime` own the published MCP configuration, connections, elicitation routing, and selected capability roots for a thread.
- Capture immutable MCP bindings for model steps and tool calls so in-flight work keeps a consistent connection set and approval authority while refreshed state is published atomically.
- Mark MCP state dirty when relevant configuration, plugins, environments, authentication, or elicitation capabilities change, then rebuild it before the next sampling request or out-of-band MCP operation.
- Separate config reloads from server invalidation: `ReloadMcpConfig` applies resolved MCP inputs, while `RefreshMcpServers` requests reinitialization from the thread's latest state.

## Testing

- Cover refreshed state visibility for existing turns, stable step bindings, current approval authority, cancelled refresh retries, resource-client reconciliation, and Apps recovery between sampling requests.

GitOrigin-RevId: 59eabb1aa8dc083426bd18ef4d3630508f376401
2026-07-23 11:27:34 +00:00
rka-oai 4462b9deef Allow disabling the multi-agent wait tool (#34887)
## What changed

- Add a default-on `features.multi_agent_v2.wait_agent_enabled` setting.
- Omit `collaboration.wait_agent` from the tool plan when the setting is disabled, independently of `features.current_time_reminder.sleep_tool`.

## Testing

- Cover configuration parsing and tool exposure with both clock sleep states.

GitOrigin-RevId: c2f34c5edd5bc191d6a5ac41d7bbb80708da1202
2026-07-23 06:28:27 +00:00
Boyang Niu 44d76c6a6d Wake sleeping threads for queued agent mail (#34852)
## Why

An idle thread with an outstanding durable sleep must resume when agent work
arrives, even when that message would normally remain queued without triggering a
turn.

## What changed

- Treat any pending mailbox message as wake-up work while a durable sleep is
  attached to the thread.
- Keep requiring `trigger_turn` for idle threads that are not durably asleep.
- Cover waking a sleeping root thread from queue-only agent mail and persisting
  the message in thread history.

GitOrigin-RevId: e361ac1b104436d4d63aebfe5e2422691a1cc139
2026-07-23 01:04:01 +00:00
Won Park 0a0a9b6c8f Disable image generation for Free-plan accounts (#34850)
## What changed

Skip registering the standalone `image_generation` tool when cached
authentication identifies the account plan as Free. Leave the existing
feature, provider capability, model modality, and authorization checks intact
for other account plans.

GitOrigin-RevId: 54d3b73dc610e508db7e281c4d7a8e72961e35a6
2026-07-23 00:56:21 +00:00
Won Park 9d82334302 Use Guardian model limits for review sessions (#34847)
## Why

Parent context-window and auto-compaction overrides may describe a different
model than the one selected for Guardian review.

## What changed

- Clear `model_context_window` and `model_auto_compact_token_limit` when the
  effective Guardian model differs from the parent model, allowing the review
  model's own limits to apply.
- Preserve those overrides when Guardian uses the same effective model.

## Testing

Add unit coverage for both model-selection paths and verify that a Guardian
rollout records the selected review model's context window.

GitOrigin-RevId: ebc399f4eba37c7f279fd0d258154c00d45e2d00
2026-07-23 00:04:37 +00:00
sayan-oai 0f9fb40fa9 Allow custom providers to opt into standalone web search (#34846)
## What changed

- Add the `supports_standalone_web_search` model-provider setting, defaulting to `false`, and preserve it in remote thread configuration.
- Enable the standalone `web.run` tool for opted-in custom Responses providers when web search is enabled and the runtime provider supports it.
- Send standalone search requests through the custom provider's endpoint and authentication.

## Testing

- Cover opt-in, default-off, disabled-search, and unsupported-provider behavior.
- Verify custom-provider search request routing and authorization through the app server.

GitOrigin-RevId: 7c5f96b0ce924ad2b9715c45bfc635e89fc39cff
2026-07-23 00:02:15 +00:00
sayan-oai 0da13c6c99 Track multi-agent mode in world state (#34845)
## Why

Multi-agent mode instructions are durable model context and need to survive
history changes without re-emitting unrelated setup hints.

## What changed

- Add a `multi_agent_mode` world-state section that diffs, retains, and restores
  the effective mode instructions.
- Persist mode changes in world-state snapshots while keeping the legacy
  `TurnContextItem.multi_agent_mode` field for reading older rollouts.
- Keep initial mode instructions after the root-agent usage hint so the active
  mode takes precedence, and bound custom mode text before snapshotting it.

## Testing

- Add snapshot coverage for mode transitions, retained history, and custom-text
  truncation.
- Verify live mode changes preserve instruction ordering, avoid duplicating the
  usage hint, and record both modes in world state.

GitOrigin-RevId: 96b9964d6e5299b07a011c559b68e3d948a49226
2026-07-22 23:57:24 +00:00
joeytrasatti-openai 400ee190c3 Add persisted thread pinning to the app server (#34840)
## What changed

- Add `isPinned` to thread responses and allow `thread/metadata/update` to pin or unpin stored threads.
- Add an `isPinned` filter to `thread/list`, including cursor-based pagination and combinations with relationship filters.
- Persist pin state in SQLite without modifying rollout files, default existing threads to unpinned, and preserve pins through reconciliation and archive transitions.

## Testing

- Cover protocol serialization, database migration and indexing, pin updates, filtered pagination, rollout reconciliation, and archive/unarchive behavior.

GitOrigin-RevId: c2eacabd6dae465e2ae6ce8a169e00740960cdb7
2026-07-22 22:54:28 +00:00
Ahmed Ibrahim d7e8f4c3dc Preserve user input when MCP startup is interrupted (#34839)
## Why

Interrupting a turn while MCP tools were still starting could abort before the
submitted user input was recorded in conversation history.

## What changed

- Build and retain the MCP tool list and tool router as part of each step
  snapshot, using the turn cancellation token during tool discovery.
- Record turn hooks and submitted input when cancellation occurs during startup
  prewarming, pre-sampling compaction, or first-step capture.
- Reuse the captured router for sampling, compaction, and prompt debugging so
  advertised and executable tools share the same request-scoped view.

## Testing

Add coverage that interrupts each affected MCP startup phase and verifies that
the user prompt remains in history before the turn-aborted marker.

GitOrigin-RevId: 6e96c6814357a55545d69b580e28f35933cfadf7
2026-07-22 22:50:13 +00:00
Ben Romano bd5b55e403 Track compaction time in turn profiles (#34835)
## What changed

- Add `compaction_ms` to turn profile facts and turn analytics events.
- Measure manual and automatic compaction as an exclusive profile phase instead of including it in surrounding idle time.
- Complete the profile and total turn duration from the same instant so their elapsed times remain consistent.

## Testing

- Cover compaction phase accounting and profile-duration consistency.
- Update turn event serialization coverage for `compaction_ms`.

GitOrigin-RevId: 50f0e1cf19789a807f9f56c2ef763c620a5d641c
2026-07-22 22:13:02 +00:00
jif 12c115d558 Reduce cloning when building Responses requests (#34825)
## Why

Preparing Responses API requests rebuilt tool definitions as a generic JSON
tree, and incremental WebSocket requests cloned their full item prefix for
comparison.

## What changed

- Serialize tool definitions into shared raw JSON that can be embedded directly
  in HTTP and WebSocket requests.
- Compare incremental request prefixes in place while still ignoring internal
  message metadata.

## Testing

- Verify raw tool JSON matches the existing value encoding.
- Preserve the serialized WebSocket request payload.

GitOrigin-RevId: 66e2921792c332e8904954dafa597f6d39abcf29
2026-07-22 21:00:01 +00:00
Adam Perry @ OpenAI bbfc3f0152 Normalize Guardian review cwd reuse keys (#34824)
## What changed

- Store the spawned session's working directory as a `PathUri` in
  `GuardianReviewSessionReuseKey` so reuse comparisons use its canonical URI
  representation.
- Assert that the reuse key contains the converted working directory.

GitOrigin-RevId: 68d440a37ebb310c610dabbb3373ed0ceff88dfa
2026-07-22 20:53:32 +00:00
jay 88eb3a2b8a Enable git attribution across Codex entry points (#34819)
## What changed

- Install the git attribution extension in the app server, MCP server, and `codex debug prompt-input` so authenticated workspace policy controls the commit and pull request attribution instructions sent to the model.
- Resolve attribution settings from the process-level ChatGPT base URL, independent of per-thread or per-tool configuration overrides.
- Treat git attribution as contextual developer content when mapping model events.

## Testing

- Cover policy fetch retries, workspace switches, rollbacks, cold thread resumes with legacy instructions, MCP tool calls, and prompt-debug output.

GitOrigin-RevId: 57d182c432c20d7c1c6c429057b6163c9f32088d
2026-07-22 20:23:35 +00:00
jiayuhuang-openai 4ebd976312 Support configurable realtime BEM channel prefixes (#34816)
## What changed

- Route realtime V3 `bemTags` handoffs using `[ANALYSIS]`, `[COMMENTARY]`, and `[FINAL]` prefixes by default.
- Add `codexResponseHandoffChannelPrefixes` to realtime start requests so clients can replace the accepted prefixes for individual channels. Each channel may accept multiple prefixes, and omitted channels retain their defaults.
- Apply the configured prefixes to both complete and streamed handoff messages, without treating empty prefixes as matches.

## Testing

- Cover request deserialization, default and custom routing, partial streamed prefixes, empty-prefix handling, and end-to-end V3 handoffs.

GitOrigin-RevId: d0aef50aa71362a5beef2c6f6d28b8633da4e529
2026-07-22 19:43:35 +00:00
pakrym-oai 08ae0fc0ce Consolidate thread startup around StartThreadOptions (#34814)
## What changed

- Add `StartThreadOptions::new` to provide the standard configuration for a new thread.
- Make `ThreadManager::start_thread` the single thread-start entry point and migrate callers from the previous convenience methods.
- Derive default environment selections when `environments` is `None`, while preserving explicit selections, including an empty list.

GitOrigin-RevId: 8977dc11aed54c5e1215a81eaed2b2cf5fc6087a
2026-07-22 19:31:54 +00:00
Adam Perry @ OpenAI 946ed315a4 Centralize SQLite connection configuration (#34808)
## What changed

- Add `SqliteConfig` to own the resolved SQLite home, runtime database paths,
  and shared read/write and read-only pool settings.
- Pass the configuration through state and local thread-store consumers,
  replacing standalone path helpers and duplicated connection setup.

GitOrigin-RevId: 7351b08da94b5c8b6c0bbe492f86aeeca0699d3c
2026-07-22 18:45:47 +00:00
Adam Perry @ OpenAI a59a419afa Use path URIs in shell approval keys (#34806)
## What changed

- Store the shell approval key's `cwd` as a `PathUri`, matching the URI representation used by approval actions.
- Extend the approval-key test to verify both the URI-based working directory and environment scoping.

## Testing

- Added `approval_key_uses_path_uri_and_includes_environment_id`.

GitOrigin-RevId: 4d7ba1a2e652f4ffe317242f52beb33513b01e6a
2026-07-22 18:41:11 +00:00
jif 66bd101fff Avoid unnecessary post-sampling token estimates (#34789)
## Why

The post-sampling trace calculated an estimated token count even when only the
always-on diagnostic and state log sinks were listening, adding unnecessary work
after every sample.

## What changed

- Move the estimate to a dedicated trace target and calculate it only when that
  event is enabled.
- Disable the target for the always-on diagnostic and state log sinks while
  keeping it available to explicitly configured subscribers.

## Testing

- Verify that the combined always-on subscribers do not enable the estimate
  event.

GitOrigin-RevId: ea572584597dda313e7719897d0ab1fdf7362745
2026-07-22 16:50:09 +00:00
Owen Lin c5779ed6bb Use the live parent history mode when forking agents (#34779)
## What changed

- Require the parent thread to be live before forking its history.
- Read the history mode and multi-agent usage hints from that live parent so
  the fork loads and filters history using the active configuration.
- Verify that forking a legacy parent keeps the child in legacy history mode.

GitOrigin-RevId: 79416b78544467a1801f21300d3fe8db4fef59df
2026-07-22 16:11:18 +00:00
thomas 65ae4c26e0 Register the MCP 2026-07-28 feature flag (#34747)
## What changed

- Add the disabled-by-default, under-development `mcp_2026_07_28` feature.
- Include the feature in the configuration schema and app-server experimental
  feature enablement API.

GitOrigin-RevId: 869835ee72f063d79ca2ecd9f2655d2efc457292
2026-07-22 12:44:31 +00:00
felixxia-oai bd9a28a839 Drop skill descriptions before omitting catalog entries (#34738)
## What changed

When even the minimum skill catalog exceeds its metadata budget, render each
included entry without its description. This allows more skill names and
locators to fit before the remaining entries are omitted.

## Testing

Add coverage for extreme budget pressure and verify both the extension and
production-turn catalogs omit descriptions while staying within budget.

GitOrigin-RevId: 72cec10e9587beabbd3d8c6d53b931deb76824e6
2026-07-22 11:53:53 +00:00
jif fd51e50540 Remove step-scoped data from extension contributors (#34734)
## What changed

- Remove the step-scoped `ExtensionData` argument from context, turn-input, and tool contributors.
- Pass the host's optional `McpResourceClient` through `ThreadStartInput` so extensions can retain session capabilities explicitly.
- Keep the MCP resource client in skills-owned session state for catalog loading and skill tools.

GitOrigin-RevId: bafa77bcd998aff408d6a396c5fd9ac268c4cce4
2026-07-22 11:17:28 +00:00
felixxia-oai 6278742c41 Preserve skill catalog entries under metadata pressure (#34732)
## Why

Long skill descriptions can consume the catalog's metadata budget before later
skills are listed, hiding otherwise usable skills from the model.

## What changed

- Reserve space for every skill's name and locator when those minimum lines fit.
- Distribute the remaining token or character budget across descriptions in
  round-robin order.
- Fall back to omitting entries only when the minimum catalog cannot fit.

## Testing

Added extension and production-turn coverage that verifies moderate budget
pressure keeps every catalog entry, shortens descriptions evenly, and avoids an
omission marker.

GitOrigin-RevId: 41971e4b47a86863b3839707b5cffcd8d83b888c
2026-07-22 11:08:38 +00:00
acrognale-oai 33d394c69e Skip Git enrichment for prewarm and Guardian turns (#34728)
## Why

Startup prewarms and Guardian review sessions do not need workspace Git metadata. Collecting it for those requests performs redundant work, while user turns still need a fresh view of the repository.

## What changed

- Add a per-session Git enrichment policy.
- Skip enrichment for startup prewarm, preview, and Guardian review turns.
- Keep fresh enrichment enabled for regular sessions and one-shot delegated turns.

## Testing

- Cover prewarm and Guardian requests omitting workspace metadata while subsequent user turns observe repository changes.
- Cover concurrent repositories and worktrees retaining distinct metadata.
- Verify concurrent enrichment is coalesced and canceled enrichment can be retried.

GitOrigin-RevId: 4a68615288bb568c75f2e487d9c0776e3e80f191
2026-07-22 10:47:51 +00:00
jif 9fc715c086 Order unified exec lifecycle events reliably (#34713)
## What changed

- Treat output-task closure as the signal that trailing output is complete, drain any remaining chunks before publishing command completion, and retain the grace period as a fallback.
- Wait for deferred network-denial classification before emitting the final command result.
- Serialize `write_stdin` interaction and completion events so an interaction that exits a process is published first, and avoid pruning processes while their terminal events are being finalized.

## Testing

- Add coverage for late output, missing output-close signals, late network denials, interaction/completion ordering, cross-platform aggregated output, and pruning during finalization.

GitOrigin-RevId: 46a118552bb1a4658b78aa5a5ca76eb8c5fba571
2026-07-22 09:29:09 +00:00
jif 516f1e2aff Rename the MCP connection manager to McpConnectionSet (#34708)
## What changed

- Use `McpConnectionSet` as the primary name for the collection of active MCP
  server connections throughout `codex-mcp` and its callers.
- Keep `McpConnectionManager` as a backward-compatible type alias.

GitOrigin-RevId: 9753838b4fd0a0b6429a9c4af0c077137616f388
2026-07-22 09:25:22 +00:00