Commit Graph
161 Commits
Author SHA1 Message Date
Anthony Tafoya f7a135ec61 Reconcile V2 executor plugins with cloud catalogs (#55301)
## What changed

- Use V2 discovery inventories for plugin MCP contributions when supplied, including empty or failed inventories; retain legacy discovery when no V2 snapshot is supplied.
- Merge plugins by canonical identity, using cloud metadata and enablement for remote plugins and attaching executor installations only when versions match. Preserve configured executor-local plugins and executor enablement when cloud data is unavailable.
- Keep installation order stable across refreshes and run merged MCP declarations on the first retained executor installation while preserving attribution for the others.
- Carry local and remote identities through selected-plugin snapshots, skill attribution, and active-plugin telemetry. Track disabled plugin IDs alongside disabled roots.
- Preserve a complete cloud catalog when a partial refresh arrives, and cap tool-description plugin attribution at 1,024 bytes while retaining all plugin associations.

## Testing

Add coverage for catalog parsing and reconciliation, enablement and discovery-mode selection, installation refresh ordering, executor-specific MCP normalization, selected-plugin snapshots, and bounded UTF-8 attribution text.

GitOrigin-RevId: a8ad352de6a2b8dc4bfd1ff6a5c62ee8ed1d1664
2026-10-19 15:15:17 +00:00
pakrym-oai ce0323f2f2 Support typed content in contextual prompt fragments (#54776)
## Why

Contextual fragments and extension-owned world-state sections currently return strings, limiting them to text content.

## What changed

- Return `ContentItem` from fragment bodies and rendering, preserving classification and attribution. Apply text markers only to `InputText`; pass other content through unchanged.
- Allow extension prompt and world-state contributions to carry typed content, and preserve media in context-window hints.
- Hash non-text world-state fragments by their serialized content and role while retaining the existing text hashing behavior.
- Add shared content-to-tool-output conversion that preserves image references, image detail, and audio.

## Testing

Add coverage for typed world-state hashing and preservation of image content and attribution. Expand code-mode output tests to cover input text, output text, file-backed and inline images, and audio.

GitOrigin-RevId: 2ed00d466a3f169b151202f3eef71e667ea64be1
2026-10-17 04:02:21 +00:00
tongzhou wang 1b6e0d8145 Track content provenance through prompt assembly and compaction (#53831)
## Why

Message roles and content classifications do not identify who supplied each context part. Source attribution needs to survive prompt assembly and history truncation, including messages that combine text from different producers.

## What changed

- Add per-part `ContentItemMetadata` for user input, developer instructions, `AGENTS.md`, skills, harness context, command hooks, and tool context with optional callable namespaces.
- Carry attribution alongside content and classifications through contextual fragments, extension prompts, hook output, and agent messages. Keep tool-provided context-window hints separate from harness text.
- Preserve the source of retained text slices during compaction and attribute generated truncation notices to the harness. Default missing historical attribution to unknown.
- Attribute host MCP calls using cached tool namespaces only when catalog revisions remain stable, without starting servers or refreshing catalogs for attribution.
- Send the new metadata to Responses Lite and strip it from requests to endpoints that do not support it.

## Testing

Extend compaction tests to cover mixed sources, missing or extra metadata entries, and discarded media. Update Responses Lite request assertions to verify endpoint-specific attribution handling.

GitOrigin-RevId: b0ebc929b7c60f3ec49d1affbf897312d63edcc8
2026-10-14 02:01:49 +00:00
Anthony Tafoya 20b4ff2f35 Enrich V2 capability discovery with resolved plugin and skill metadata (#53763)
## What changed

- Return qualified skill names such as `team:deploy` and parsed `dependencies` and `policy` instead of a separate namespace and raw metadata file. Share `agents/openai.yaml` parsing and validation between skill loaders and executor discovery.
- Include each standalone skill's search-root ID and path in `discoveryRoot`, while accepting responses from older executors that omit it.
- Add plugin `displayName`, falling back to the manifest name or installation directory.
- Materialize inline MCP declarations in `mcpConfig` with the manifest as their source path, and report incomplete discovery when declared MCP or Apps configuration files have no returned contents.

## Testing

Add coverage for JSON round trips, omitted `discoveryRoot`, optional and invalid skill metadata, qualified names, display-name fallbacks, and inline MCP configuration. Verify standalone search roots survive cached discovery and sandboxed RPC responses.

GitOrigin-RevId: 4124972444ad7c2fb8c6418acbf9a589d797fad2
2026-10-13 20:40:38 +00:00
Anthony Tafoya cf0ed4c600 Acquire and cache opt-in executor capability inventories via V2 (#53736)
## What changed

- Add the disabled-by-default `executor_capability_discovery_v2` feature to acquire `capabilities/discoverV2` inventories from supporting remote executors using each selected environment's working directory and sandbox.
- Share snapshots with MCP and world-state contributors and skill providers without activating discovered plugins or publishing discovered skills to the model.
- Cache inventories by environment instance, executor session, working directory, and sandbox. Coalesce concurrent requests, refresh the cache at turn start, retry transient failures, and refresh MCP projections after recovery.
- Resolve the system skills root in the executor and pass it explicitly to capability location discovery.

## Testing

Add coverage for feature gating, unsupported executors, inert inventories, cache reuse and eviction, turn refreshes, failure recovery, executor replacement, and system skill locations.

GitOrigin-RevId: 1bdca28fa7f17622b27e2713ab3dcc0c13426889
2026-10-13 18:15:50 +00:00
Adam Perry @ OpenAI 806d2d1cc1 Remove per-content source attribution metadata (#52329)
## What changed

- Remove `ContentItemMetadata`, provenance and namespace types, and the `content_item_metadata` protocol field while retaining `content_item_kinds` classifications.
- Simplify context fragments, extension contributions, and hook output to carry content without source attribution. Return MCP tool results directly without an attribution wrapper.
- Render context-window hints together in a single text fragment.

## Testing

Update Responses and Responses Lite request assertions to expect content-kind classifications without per-content attribution, and adapt context, hook, and extension tests to the simplified types.

GitOrigin-RevId: 201894cd4c93c6d4b4c41c845e7a9af1dd3c95ca
2026-10-08 23:57:36 +00:00
Alex Zamoshchin c9e0fffbbc Enable parallel execution for read-only tools (#52245)
## Why

Tools that do not opt into parallel execution acquire an exclusive scheduler lock, serializing independent reads.

## What changed

Allow skill listing and reading, memory listing, reading and searching, conversation history search and reading, plugin listing, agent listing, current time, remaining context, and `get_goal` to run in parallel. Keep goal creation and updates serialized.

## Testing

Add `cloud_skill_reads_run_in_parallel`, which uses a barrier to require two cloud skill reads to overlap and verifies that each returns the expected resource contents.

GitOrigin-RevId: 97b5634f1f84be2ba092be89e6e9c634ca75c07b
2026-10-08 19:29:47 +00:00
tongzhou wang ad54b75574 Track source attribution for individual context parts (#52195)
## What changed

- Add `ContentItemMetadata` to distinguish user input, developer instructions, `AGENTS.md`, skills, harness context, and tool-supplied context independently of message role and content classification.
- Keep attribution attached to content through prompt assembly, history processing, and hook-output spilling. Distinguish command hooks from MCP hooks and retain tool namespaces when known.
- Treat missing historical attribution as unknown and send per-content metadata only to Responses Lite.

## Testing

Update request and history assertions to cover attribution persistence and Responses Lite metadata handling.

GitOrigin-RevId: 24e56005f5bdb98bf82f2b3943f29c4c81f3d587
2026-10-08 17:56:54 +00:00
Adam Perry @ OpenAI c9870d0157 Use PathUri for skill identity and path matching (#51482)
## Why

Skill selection and disabled-path matching need to recognize equivalent Windows paths across case and separator differences while preserving literal spaces, `%`, and `#` in native filenames.

## What changed

- Carry `PathUri` through skill metadata, discovery mappings, filesystem lookups, disabled-path sets, and invocation analytics.
- Match linked skill mentions and TUI selections by parsed path identity.
- Use `LegacyAppPathString` for app-server skill paths and regenerate schemas and SDK types, preserving native path strings on the wire.
- Preserve native catalog display paths and ordering. Keep host filesystem probes and configuration writes restricted to host-compatible paths.

## Testing

Add regression tests for Windows path identity and disabled-skill matching, literal filename characters, catalog resource reads, native path ordering, and foreign-path analytics normalization.

GitOrigin-RevId: c212916dc9fef0e7616542726e107f0d6a313267
2026-10-06 21:29:35 +00:00
Adam Perry @ OpenAI 42312d4ff4 Enforce required environment skills before model inference (#51157)
## What changed

- Accept per-environment `skills.required` lists through `environment/add` and `environments.toml`.
- Before inference, require each named skill to be enabled and supplied by the selected environment. Missing skills fail the turn before a model request, including when environment setup fails.
- Stop applying an environment's requirements after it is deselected. Exempt isolated Guardian reviewers, which have no skill catalog.

## Testing

Add tests for available, missing, disabled, and wrong-environment skills, failed environment setup, recovery after deselection, and propagation of TOML requirements to the correct environment.

GitOrigin-RevId: 302ced4774697d2f414f745fe0535e71f747eda2
2026-10-05 18:54:17 +00:00
pakrym-oai 8571b9eaa4 Install full context in compaction replacement history (#51117)
## Why

Manual, pre-turn, and post-turn compaction previously deferred rebuilding initial context until the next turn. Their saved replacement history could omit refreshed instructions and skill catalogs.

## What changed

- Install initial context during local and remote compaction, before the last real user message or summary, keeping the compaction summary last.
- Persist matching full world-state and turn-context baselines with every replacement instead of retaining only partial extension metadata.
- Build replacement context and recompute token usage for the current model even when the previous model performs compaction. Reuse that captured context for eligible current-model retries.
- Capture manual compaction context centrally and propagate preparation errors to the task runner.

## Testing

Update assertions and snapshots for context ordering, refreshed instructions, resume and fork history, current-model checkpoint baselines, and cloud skill catalog retention across compaction and resume.

GitOrigin-RevId: 8572a76c31c6884a3c580b3bbb350d49af876bd1
2026-10-05 16:10:27 +00:00
sayan-oai 9561a34531 Keep plugin and skill snapshots scoped to each step (#50050)
## Why

Environment changes and overlapping catalog refreshes can cause a step to use another step's selected plugins or executor skills. Model context and skill tools need to agree on the capabilities captured for that step.

## What changed

- Resolve selected plugins from each step's ready capability roots, honoring disabled plugins, and retain the snapshot with the step's extension data.
- Use that snapshot for skill attribution in World State and the first step's active-plugin telemetry.
- Return executor skill catalogs directly from refresh so overlapping requests render their own results while retaining thread caches.

## Testing

Add regression coverage for overlapping executor catalogs and environment changes during step preparation. Verify that `skills.list` retains the original step's skills, the next step advertises the new skills, and disabled executor and plugin skills remain absent from model context.

GitOrigin-RevId: 37d3cbe8c7e062fd06290aa985ada4d3b3462d43
2026-10-01 17:44:52 +00:00
Adam Perry @ OpenAI da2e174a66 Scope extension filesystem access to callback permissions (#49898)
## Why

Executor skill reads need to use the environments and permissions captured for the current step, including newly approved grants. Cached continuation pages must remain bound to the filesystem and permissions that produced them.

## What changed

- Supply borrowed `EnvironmentAccess` to turn-input contributors and extension tools, with session and originating-turn grants already applied.
- Require executor skill reads to use filesystem access from an environment available to the callback, and replace separate optional read contexts with `SkillReadContext`.
- Bind skill pagination snapshots to `EnvironmentAccessKey`, which identifies the source filesystem and callback permissions.
- Route image-generation filesystem operations through the same accessor.

## Testing

Extend skill-read integration coverage to verify that a denied reference becomes readable after a turn-scoped grant is captured by the next model step. Update Windows coverage to verify that restricted reads fail when the executor cannot enforce a sandbox, and add a test requiring turn environment selection before exposing capability roots.

GitOrigin-RevId: 825b49302319965835d61c77fc5bcdb951e10543
2026-10-01 06:29:03 +00:00
pakrym-oai b44ca872b1 Persist world-state snapshots alongside rendered context (#49847)
## What changed

Return snapshots and model-visible fragments together from world-state rendering, and use those snapshots for history baselines, step updates, compaction, and new context windows.

Update `WorldStateSectionContribution` so its renderer returns an optional replacement snapshot and an optional fragment. A missing snapshot preserves stored section state, even when retained history requires a fresh render; a null snapshot removes the section. Derive step patches from the history baseline rather than a separately captured live world state.

Adapt skills and Git attribution contributors to the new contract.

## Testing

Add coverage for skipped snapshot updates, explicit section removal, changed-state patch replay, unchanged-state deduplication, and matching live and persisted checkpoints when starting a new context window.

GitOrigin-RevId: af71606a278358ae937853cb76e8a019a501408e
2026-10-01 03:38:50 +00:00
vkg-oai 0f8df3d214 Move shadow skill ranking off the turn preparation path (#49812)
## Why

Shadow skill selection runs synchronously during turn input construction, making turn preparation wait for experimental rankings.

## What changed

Run ranking on background blocking workers with a process-wide limit of two concurrent rankings. Wait for capacity in the background and hold each slot until its blocking worker finishes, even if the async evaluation is cancelled.

Freeze catalog, query, and relevance history at turn start. Record eligible invocations immediately for future turns, then score queued observations against the frozen rankings once they are ready, including observations queued before the turn state is dropped.

## Testing

Add tests for turn preparation and history updates while ranking is blocked, deferred invocation scoring against each turn's history, and shared ranking capacity that remains occupied until cancelled workers finish. Update existing metric assertions to collect asynchronous observations across snapshots.

GitOrigin-RevId: f7776e9aaeb45d3a792de167abbe35f7ec5a513e
2026-10-01 01:49:59 +00:00
jif 0a76a2520f Export skill invocation events through OpenTelemetry (#49689)
## What changed

Emit `codex.skill_invocation` logs for explicit skill injection and detected implicit use, including skill-tool reads. Include skill name, invocation type, conversation, turn, model, and client metadata, plus scope, plugin ID, and ChatGPT user and account IDs when available.

Exclude skill contents, resource paths, email addresses, prompts, and tool payloads. Use the existing OTEL log exporter without an additional opt-in; `otel.exporter = "none"` continues to export no logs.

Document the event fields, exporter configuration, and detection semantics in `codex-rs/otel/README.md`.

GitOrigin-RevId: 0f9c492c3f224455f3de3660be89692d776c1951
2026-09-30 17:45:25 +00:00
Anthony Tafoya 13f580ef09 Deduplicate cloud and executor skill listings before budgeting (#49127)
## Why

Skills available from both cloud and executor providers consume duplicate catalog space, leaving less room for descriptions of unique skills.

## What changed

- Prefer model-visible cloud skills over executor listings with the same `plugin:skill` name before allocating the metadata budget.
- Preserve aliases from the full visible executor inventory so filtering duplicates does not renumber remaining packages. Continue choosing full locators when they fit the budget better.
- Keep executor packages available through `skills.list` and `skills.read`, and retain executor listings when cloud skills are disabled.

## Testing

Add regression tests and a model-request snapshot covering description retention, stable aliases across turns, executor availability changes, package reads, and full-locator selection after deduplication.

GitOrigin-RevId: 477e4cf16b8968dbe61438dd5af1eca9cbfa78c8
2026-09-29 01:57:57 +00:00
Matthew Zeng 456212ca21 Use explicit histogram buckets for tool and skill context metrics (#48819)
## What changed

- Give tool fragment sizes and namespace counts logarithmic boundaries through 32,768.
- Give enabled and kept skill counts integer boundaries from 0 through 512, removed description characters logarithmic boundaries through 131,072, and skill truncation flags boundaries at 0 and 1.
- Record these metrics with explicit boundaries, separating zero values and retaining overflow buckets above each range.

## Testing

Extend tool and skill metric tests to assert explicit boundaries. Add an export test covering zero, one, the upper boundary, and overflow, including bucket counts, ordering, and sample totals.

GitOrigin-RevId: 9fbf3acbe352d7c74f3ad11f0b690148c083a6ef
2026-09-28 00:05:54 +00:00
vkg-oai e72da2b538 Stabilize skill catalogs across executor availability changes (#48353)
## Why

Changes in executor availability can alter cloud skill catalog rendering under a shared budget and repeat an unchanged executor catalog when it reconnects.

## What changed

- Cap cloud skills at three quarters of the metadata budget initially, leaving the remaining budget and unused cloud allowance for filesystem skills. Reduce the cloud cap only when doing so allows all skill entries to fit; description truncation alone does not trigger rebalancing.
- Persist the allocation across disconnects, compaction, and thread resume. Recompute it when the cloud inventory or total budget changes.
- Emit a short availability update when an unchanged selected-environment catalog returns and its full text remains in history. Reinject the full catalog when that text is missing.

## Testing

Add coverage for allocation stability, catalog and budget changes, omission handling, and full catalog reinjection after compaction and resume. Extend the selected capability stack test to verify reconnection avoids repeating the catalog.

GitOrigin-RevId: 839fe98024e00c4082a46e6a40675e15fd2c16c6
2026-09-26 04:19:46 +00:00
sayan-oai e8098eb406 Separate selected plugin identities from MCP contributions (#47939)
## Why

Plugin identity and skill ownership need to be available independently of MCP data, including for plugins without servers or connectors. Hosted plugins have no executor root and should not own executor skills.

## What changed

- Add `McpServerContributor::selected_plugins` to declare executor plugin identities with deferred MCP data. Let the host assign server attribution and precedence across contributors.
- Represent hosted connectors with `HostedPluginConnectors` and make `SelectedPluginIdentity::selected_root_id` optional instead of assigning synthetic roots to hosted plugins.
- Resolve executor contributions before reading hosted connectors, and clear stale hosted state on account changes before executor loading can wait.
- Preserve connector declarations from disabled plugins so Apps can hide those connectors.

## Testing

Add coverage for account changes while executor loading is pending. Update tests to verify hosted identities have no executor root and disabled plugins hide their connectors even when plugin IDs differ from root IDs.

GitOrigin-RevId: e7f511264c5874c1b56d5f3e4a46820724046db0
2026-09-24 21:03:50 +00:00
Henry Levy 8edfca892d Reuse cloud skill catalogs across turns until invalidated (#47350)
## Why

Cloud skill discovery ran on every turn, even when the Apps resource generation and authentication scope were unchanged.

## What changed

- Reuse warning-free catalogs, including empty catalogs, while their authentication scope and Apps resource cache key remain current.
- Refresh after invalidation and retry failed or partial discovery on the next turn.
- Reject discovery results if the Apps resource cache key changes during discovery.

## Testing

Update lifecycle tests to verify cached turns avoid catalog listings and skill reads, while connection and authentication changes trigger refreshes. Update the referenced-resource test to verify partial discovery retries on the next turn without an explicit MCP refresh.

GitOrigin-RevId: ac47d5202d76703c4946afef3b1a48eaf05b54c2
2026-09-22 19:58:52 +00:00
Eddie Chen 6a601c00da Preserve originating turn metadata for yielded skill calls (#47106)
## Why

Code Mode callbacks can retain skill tools after another turn becomes active. Looking up the active turn when a delayed skill invocation completes can lose access to the originating turn's metrics and metadata.

## What changed

- Capture and retain the originating `SkillTurnMetrics` when constructing skill analytics.
- Expose turn analytics metadata through the turn extension store and carry it into skill telemetry.
- Add optional turn metadata to `TrackEventsContext` and pass it from app, skill, and plugin usage call sites.

## Testing

Add a regression test that yields a Code Mode cell, starts another turn, and releases both skill reads through a shared barrier. Assert that each `skill_invocation` event retains its originating `turn_id`.

GitOrigin-RevId: aa629dd5f84563d558aa6a86bfbdfc356f1b98e4
2026-09-21 20:34:46 +00:00
Anthony Tafoya f3037cafd3 Refresh host-supplied cloud skills at turn startup (#47074)
## What changed

- Replace the built-in orchestrator skill provider with a host-supplied cloud provider, using `cloud` authority and `c` root aliases in skill tools and prompts.
- Add `cloud.skills.enabled`, defaulting to true but requiring a supplied provider. Keep `orchestrator.skills` accepted as a legacy no-op setting, and remove the app server's automatic orchestrator provider registration.
- Discover cloud skills once per turn during cancellable startup. Serve catalog snapshots to prompts and tools without triggering discovery or retries.
- Scope cached catalogs and resources to authentication state. Preserve them on refresh failures within the same scope, invalidate them when that scope changes, and reject late discovery or read results from replaced cache generations.
- Reuse executor discovery catalogs only when successful discovery inputs and bundled-skill settings match.

## Testing

Add coverage for per-turn refresh, skill removal, same-account failures and reconnects, authentication changes, and stale in-flight results. Update configuration and tool tests for cloud authority, provider gating, and resource reads without an executor.

GitOrigin-RevId: c57e38ec7dcc880c0d4c8a2995adfd3e28301079
2026-09-21 17:28:14 +00:00
Adam Perry @ OpenAI 47fc8d661e Route skill discovery and loading through EnvironmentAccess (#46293)
## What changed

Use `EnvironmentAccess` for skill discovery, environment skill loading, and plugin namespace resolution, replacing direct `ExecutorFileSystem` calls with explicit `None` sandbox arguments.

Wrap existing host and executor filesystems with `FileSystemEnvironmentAccessor::unrestricted` to preserve their current access behavior. Update discovery, loading, and namespace tests to use the same adapter.

GitOrigin-RevId: d68168cc0ee19dc303ef4ee726cf1de1a8db81e0
2026-09-17 19:52:03 +00:00
Adam Perry @ OpenAI a4ee536f01 Route filesystem reads and writes by their own sandbox permissions (#46122)
## Why

Filesystem reads previously required a sandbox whenever writes were restricted, even with full-disk read permission. This made permitted reads depend on sandbox availability.

## What changed

- Select sandboxing independently for reads and writes, allowing full-disk reads directly while keeping restricted operations sandboxed.
- Use the executor's path convention when evaluating full-disk access, including `:slash_tmp` denials.
- Apply read-specific checks to capability discovery and skill resource reads, allowing unrestricted reads on executors without sandboxed discovery support.

## Testing

Add regression tests for direct read APIs, restricted writes and reads, executor-specific permission rules, and capability discovery without sandbox support. Update Windows tests to verify sandbox enforcement through writes.

GitOrigin-RevId: a03844bd1f0ea583bb54326683cddbfdd050119f
2026-09-17 05:09:03 +00:00
Eric Traut 36b84c81ec Suppress warnings when skill descriptions are shortened (#46070)
## What changed

Stop emitting a warning when skill descriptions are shortened to fit the skills context budget. Continue warning when skills are omitted from the model-visible list.

Update the app-server warning test to set `skills.max_context_tokens` to `1_000` and expect the omitted-skills warning. Remove assertions and tests for the description-shortening warning.

GitOrigin-RevId: feac10b3d159d73648a786120d477b7fe4e082d7
2026-09-17 01:28:24 +00:00
Henry Levy 29e6bc814e Preserve orchestrator skill caches across MCP runtime updates (#46033)
## Why

Orchestrator skill caches were tied to the entire MCP connection set, so runtime updates could discard cached catalogs and contents even when the Apps connection was reused. Plugin changes still need to refresh those resources without requiring a reconnect.

## What changed

Key orchestrator skill caches by the Apps server connection and an explicit invalidation generation. Preserve cached resources while that identity stays unchanged, and invalidate them on MCP runtime refresh requests or configuration refreshes. Replacing the connection also changes the cache identity.

## Testing

Extend the production-turn integration test to verify cache reuse across ordinary turns and approval-policy changes, refreshed skill contents after plugin invalidation without reconnecting, and fresh resource requests after a forced reconnect.

GitOrigin-RevId: 43ce059972369c4c735742e8224b60a692244b02
2026-09-16 21:41:23 +00:00
vkg-oai fc2ea82e7e Allow callers to disable executor skills per environment (#46015)
## Why

Executor skill discovery enabled every discovered skill, with no provider option for callers to disable selected skills.

## What changed

Add `ExecutorSkillProvider::with_disabled_skill_paths` to accept `SKILL.md` paths grouped by environment ID. Mark matching catalog entries disabled in both discovery routes so they are omitted from model context. Skills remain enabled by default, and paths configured for another environment have no effect.

## Testing

Extend catalog tests to cover default enablement, matching paths, and environment isolation across both discovery routes. Add a model-context snapshot test confirming that an active skill appears while a disabled skill is omitted.

GitOrigin-RevId: 7426eb94566db2bcf7e49a5dee28673ab743ad25
2026-09-16 20:05:37 +00:00
iceweasel-oai d4e11a9b97 Separate executor sandbox selection from Windows sandbox levels (#45730)
## Why

MXC is a sandbox implementation, not a restricted-token sandbox level. Executor requests need to represent that choice separately from `WindowsSandboxLevel`.

## What changed

- Introduce `WindowsSandboxSelection` for executor sandbox contexts and remove `Mxc` from `WindowsSandboxLevel`.
- Preserve the `windowsSandboxLevel` wire field and its serialized values for compatibility.
- Share sandbox selection between executor process launches and filesystem helpers, and use the new selection in capability discovery and skill reads.
- Disable Windows sandbox selection for executor paths that do not use Windows path conventions.

## Testing

Extend coverage for MXC wire serialization, Windows skill-read sandbox checks, and capability discovery with distinct permissions. Exercise remote filesystem write restrictions with both restricted-token and MXC sandboxes, including rejection when native MXC is unavailable.

GitOrigin-RevId: 266211377bcb138a0dc75861e9ff2225fa37a53d
2026-09-15 16:40:06 +00:00
Charlie Marsh f3803587c9 Share tool output schemas and defer MCP envelope construction (#45439)
## Why

MCP tool parsing eagerly cloned structured output schemas and built full call-result envelopes. Cloning tool definitions also copied their output-schema JSON, even before a consumer needed it.

## What changed

- Introduce `ToolOutputSchema` with immutable `Arc` storage so tool definitions share output schemas when cloned.
- Retain MCP structured output schemas and materialize the call-result envelope only when JSON is requested.
- Update code-mode consumers and schema mutation sites to materialize JSON explicitly, reusing uniquely owned storage when possible.
- Move structured content into the MCP envelope without an extra clone, preserving property order.

## Testing

Add tests for JSON preservation, mutation isolation, equality between lazy and materialized schemas, reuse of uniquely owned storage, and equivalent code-mode definitions.

GitOrigin-RevId: e98ba4c2f0efedc99f7cbc7bba206cc63a3bd8f4
2026-09-14 15:05:16 +00:00
Abhinav 8ff4aa8ee4 Use captured step model settings for extension context (#44202)
## Why

Model switches within a turn can leave extension context using stale model metadata. Skill catalogs, context windows, and metric attribution need to match the model captured for each sampling step.

## What changed

- Pass captured model metadata and step-specific telemetry to world-state contributors.
- Use that metadata for skill catalog budgets and usage instructions, preserving configured budget overrides.
- Supply the captured model's usable context window to turn-context contributors, including when rebuilding context.

## Testing

Add regression coverage for model switches during skill discovery, catalog budgets and metric attribution, and extension context windows after `new_context` and retained-step context rebuilds.

GitOrigin-RevId: c51b40b739a380b9767a54ea8c8ee6de45bf01c8
2026-09-09 16:48:49 +00:00
felixxia-oai 2bba3a29a0 Use explicit histogram buckets for Guardian context metrics (#44181)
## Why

Guardian request and section cost distributions need shared bucket boundaries across synchronous and asynchronous reviewers so their measurements align.

## What changed

- Add `histogram_with_boundaries` to session telemetry and extension metrics while preserving session attribution.
- Use shared request-token buckets up to 2,000,000 tokens and section-cost buckets up to 16,777,216 for both review paths.

## Testing

Extend telemetry tests to verify explicit bucket boundaries and sample counts, and Guardian integration coverage to check exported request and section metric bounds for both synchronous and asynchronous reviews.

GitOrigin-RevId: e34f6972b3418bac3b061939b62eeeccaec5a299
2026-09-09 15:48:39 +00:00
jif 50fffd5ed3 Refresh plugin skills after out-of-process version changes (#42284)
## Why

Another process sharing `CODEX_HOME` can replace an installed plugin version
without invalidating the current process's caches. This could leave skill
listings tied to stale plugin paths and retain obsolete plugin generations.

## What changed

- Reject cached plugin loads when their roots no longer match the active
  installation, so skills are reloaded after upgrades or rollbacks.
- Keep the 32 most recently used configuration-based skill snapshots while
  allowing callers to continue using snapshots that have been evicted.

## Testing

- Cover external plugin upgrades and rollbacks through `skills/list`, including
  a subsequent warm-cache read.
- Cover cache eviction, reuse, and the lifetime of caller-held snapshots.

GitOrigin-RevId: ca00f9539c01461e3945d340bf63f8436665220b
2026-09-02 14:14:44 +00:00
Adam Perry @ OpenAI 81e180044d Scope extension capabilities to invocation lifetimes (#41020)
## What changed

- Add callback lifetimes to extension `ToolCall`, `ToolEnvironment`, turn-input context, and skill-read request types.
- Require extension tool executors to handle calls for any invocation lifetime and tie their returned futures to that lifetime.
- Update built-in extensions, adapters, and tests to use the lifetime-scoped APIs without retaining host-owned environment capabilities.

GitOrigin-RevId: aca9ce3a1b6870df550bc84ec9308cc65533f161
2026-08-27 02:00:31 +00:00
xl-openai 7c3747941a Reduce skill catalog prompts with path aliases (#41011)
## Why

Repeated skill locator roots can make catalog prompts larger even when the
catalog fits within its metadata budget.

## What changed

Evaluate aliased catalogs regardless of budget pressure and select them when
they preserve skill inclusion and description content while reducing prompt
size. Apply the same selection to combined host, executor, and orchestrator
catalogs.

## Testing

Add coverage for alias selection in full host-only catalogs and update skill,
symlink, and orchestrator resource tests to expect shortened locators and their
root mappings.

GitOrigin-RevId: ccedd3f3f6c7efaa1e5177a2d8c3d81d59239f52
2026-08-27 01:22:23 +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
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
hefuc-oai 5ce04789fb Honor managed plugin disablement for executor capabilities (#40640)
## Why

Selected executor plugin roots could still expose capabilities when managed
requirements disabled the `plugins` feature.

## What changed

- Suppress MCP servers, skills, apps, and connectors from selected executor
  plugins when `plugins` is disabled, while preserving the selected-root
  identity used to filter those capabilities.
- Apply the policy consistently to direct selected-root discovery and batched
  executor capability discovery.

## Testing

- Add contributor-level and app-server coverage that verifies disabled plugin
  capabilities are absent and their MCP servers never start.

GitOrigin-RevId: e30bd8a936c4b0e083908b89a65f4d02e01747de
2026-08-25 16:47:13 +00:00
Alex Zamoshchin 304c8de4c6 Allow extensions to skip host skill discovery (#40631)
## What changed

- Add the under-development `skip_host_skill_discovery` feature to bypass host
  skill snapshots during session warmup and turn setup.
- Let skill invocation contributors declare whether they require host-owned
  skills. Preserve host discovery when no contributors are registered or any
  contributor requires it.
- Keep executor and orchestrator skill catalogs and instruction loading
  available when host discovery is skipped.

## Testing

- Cover executor-only, orchestrator, mixed-contributor, and legacy host-skill
  behavior.

GitOrigin-RevId: 48169403090b234e1a304c6523633fd867df454a
2026-08-25 16:00:02 +00:00
Tamir Duberstein c941572917 Honor response budgets when reading skill resources (#40491)
## Why

`skills.read` paginated resources against a fixed response limit, so a page could exceed a smaller tool-call response budget.

## What changed

- Size each serialized `skills.read` page to the current call's response budget, accounting for JSON escaping and UTF-8 boundaries.
- Cache one bounded executor-resource snapshot per thread so cursor continuations return consistent contents without rereading the file. Replace the snapshot when another resource is read, and reject stale cursors after a reread.
- Tie cached executor snapshots to the selected environment and sandbox context without keeping the environment alive.

## Testing

- Cover budget-constrained reads, escaped and multibyte contents, multi-page reconstruction, changed resources, snapshot eviction, and expired turn-scoped access.

GitOrigin-RevId: e515935b82771b3f29b3fb91ea7a7a72e19a65ee
2026-08-24 23:00:57 +00:00
Tamir Duberstein a25e986323 Honor response budgets when listing skills (#40413)
## What changed

- Build `skills.list` pages against the current tool-call response-byte budget.
- Skip entries that cannot fit, preserve pagination across retained entries, and report oversized metadata without repeating the warning for the same budget.
- Keep existing cursors compatible while encoding the response budget in new cursors.

## Testing

- Cover direct and Code Mode listing, budget changes between pages, provider and omission warnings, oversized entries, legacy cursors, and budgets too small to return discovery warnings.

GitOrigin-RevId: 4b34cd2714e4849fe9155ce3f0021b218763c395
2026-08-24 14:54:31 +00:00
Tamir Duberstein e3609f2d02 Expose invocation sources to extension tools (#40382)
## What changed

- Add `ToolCallSource` to extension `ToolCall`s and propagate direct or Code Mode
  invocation metadata, including the runtime cell and nested tool-call IDs.
- Add `ToolCall::response_byte_budget` so extension tools can bound direct-call
  responses by the host truncation policy while allowing Code Mode responses up
  to the tool's own limit.
- Share the skills extension's response-size limit between its `list` and `read`
  tools.

## Testing

- Verify that Code Mode source metadata is preserved when core constructs an
  extension tool call.

GitOrigin-RevId: b10aef01be30d35b67bee3af24a1a4175c9ee112
2026-08-24 11:02:14 +00:00
pakrym-oai 422239eb4b Classify contextual fragments with content kinds (#40180)
## What changed

- Require each `ContextualUserFragment` to provide a stable `<feature>.<name>`
  `ContentItemKind`.
- Add `AnnotatedContent` and `RenderedFragment` so rendered text, its role, and
  its classification can travel together to API boundaries.
- Derive extension-owned world-state classifications from the extension ID and
  keep the skills catalog classification with its fragment implementation.

## Testing

- Verify that an extension-owned world-state section renders with an
  `<extension-id>.instructions` content kind.

GitOrigin-RevId: e46b74a0bb41e0b6112667c9d36bc9e7f2714451
2026-08-23 03:44:21 +00:00
pakrym-oai 546eed94ac Require content kinds for extension prompt fragments (#40177)
## What changed

- Require each `PromptFragment` to include a producer-owned `ContentItemKind`.
- Re-export `ContentItemKind` from `codex_extension_api` and assign stable kinds to skills, memories, examples, and test contributors.
- Limit prompt slots to developer policy and developer capabilities, and combine contributed fragments into a single developer update.

## Testing

- Update extension registry tests to verify fragment ordering with the new content-kind metadata.

GitOrigin-RevId: ef8624a9b4ae9dce025bca9c0954d91dd8075e85
2026-08-23 03:12:58 +00:00
pakrym-oai e3e5ad2847 Harden unsandboxed patch filesystem access (#39659)
## Why

An `apply_patch` path can be replaced with a symlink after verification, allowing an unsandboxed patch operation to reach a different file than the one that was approved.

## What changed

- Add `follow_symlinks` options to executor filesystem reads, writes, metadata lookups, directory creation, and removal, including the corresponding `followSymlinks` protocol fields.
- Implement no-follow filesystem operations on Unix and Windows that reject links in any path component and restrict file access to regular files.
- Run `apply_patch` with symlink traversal disabled when an otherwise-required sandbox is bypassed, while retaining the existing follow-symlink default for standalone callers.

## Testing

- Cover leaf and ancestor symlinks across patch add, update, delete, and move operations, including a path swap after verification.
- Exercise local and remote no-follow filesystem behavior, concurrent directory creation, special-file rejection, and Windows reparse points.

GitOrigin-RevId: 43fd479084891493ce13564fbd894b98f329c6dd
2026-08-20 08:10:08 +00:00
Adam Perry @ OpenAI f1e06b3865 Stream executor capability and skill file reads (#39620)
## What changed

- Use `ExecutorFileSystem::read_file_stream` for capability files and executor skill resources regardless of whether filesystem sandboxing is active.
- Enforce per-file and bundle size limits incrementally as chunks arrive.
- Preserve the existing error for Windows resources when the required filesystem sandbox is unavailable.

GitOrigin-RevId: 1a40602a8f913588ec9a6ad7edd2d62eb82436e1
2026-08-20 06:43:50 +00:00
rphilizaire-openai 280d56b1d8 Attribute executor skill invocations to plugins (#39309)
## What changed

- Carry selected plugin identities from MCP discovery into per-turn extension data, limited to capability roots that are ready for the turn.
- Annotate executor skill catalog entries with the matching plugin ID and `user` scope.
- Include that attribution in `skill_invocation` analytics for both explicit and implicit executor skill invocations.

## Testing

- Extend the executor skills app-server test to verify plugin ID, skill scope, and invocation type for explicit and implicit analytics events.

GitOrigin-RevId: 9345600ae29bcb8eb894fffd2d6f7b05f22ebf74
2026-08-18 22:53:31 +00:00
Dylan Hurd e7e13c68e2 Add Guardian v2 approval review metrics (#39224)
## What changed

- Pass metrics bound to the active turn's model to approval-review contributors.
- Add counter support to `ExtensionMetrics` while preserving host session attribution.
- Record Guardian v2 tool-call score lag and count stale-score fallbacks with the `score_lag` reason.

## Testing

- Cover counter forwarding and session metadata tags.
- Verify Guardian v2 metric samples for current and stale scores.

GitOrigin-RevId: 0ec34736d29d2e37ed217a27e773b4f6230ae38a
2026-08-18 16:27:54 +00:00
felixxia-oai d24507a59b Remove skill model delegation support (#39068)
## What changed

- Stop parsing and exposing the `model` field from skill frontmatter.
- Remove the skill model delegation types and instruction generation.

GitOrigin-RevId: 01198c68c095da5062e8abec417cc3539099d9d5
2026-08-17 20:19:04 +00:00
jif 21cfd369ef Add task context to shadow skill selection (#39008)
## Why

Short continuation prompts such as `continue` do not contain enough context for
skill selection on their own.

## What changed

- Add the `task_context_fusion_v1` shadow selector, combining the current request
  with up to two prior substantive requests and recently relevant skills.
- Record explicit skill intent and successful skill invocations for future turns,
  while excluding same-turn observations from predictions.
- Bound retained requests, augmented queries, and skill history, including safe
  truncation at UTF-8 character boundaries.

## Testing

Add unit and extension tests for continuation prompts, explicit intent, turn
isolation, cold thread state, bounded history, and unchanged control selectors.

GitOrigin-RevId: 72eca3f0d64620a0d24e95d5675f126ac982f8c7
2026-08-17 13:29:57 +00:00