Commit Graph

9299 Commits

Author SHA1 Message Date
Adam Perry @ OpenAI
ccee70f813 Remove repository-local Codex skills (#38635)
## What changed

- Remove the `codex-bug` issue-triage skill.
- Remove the `codex-issue-digest` skill, including its collector, agent metadata, and tests.
- Remove the `pushing-ci-changes` skill.

GitOrigin-RevId: 90f69a32b09456271344d49ebe5aa3eed5a8c808
2026-08-14 20:02:03 +00:00
thomas
6bed213411 Add MCP protocol discovery metrics (#38634)
## What changed

- Record a counter and duration for MCP client protocol discovery.
- Tag observations with the configured `legacy` or `auto` mode and classify
  outcomes as `modern`, `legacy`, or `failure`.

GitOrigin-RevId: f348e0d900b437d5a9fef1c33cdb1314f2421785
2026-08-14 19:57:04 +00:00
Adam Perry @ OpenAI
6595071e65 Remove the gRPC code-mode open session limit (#38630)
## What changed

Allow the gRPC code-mode host to register more than
`MAX_IN_FLIGHT_REQUESTS` open sessions. Existing limits on in-flight
requests, control requests, and active cells remain unchanged.

GitOrigin-RevId: 126c5088868e7783f8592f3f9250f5c8573df51f
2026-08-14 19:26:36 +00:00
felixxia-oai
baab1705c6 Make Guardian v2 risk classification configurable (#38628)
## What changed

- Allow `features.guardianv2` to remain a boolean toggle or specify classifier instructions, the review threshold, reasoning effort, and action and instruction token limits.
- Add transcript controls for included sources, per-entry and total token budgets, and the number of recent non-user entries.
- Validate configured ranges and relationships, expose them in the generated config schema, and apply the resolved settings throughout Guardian v2 classification and approval review.

## Testing

- Cover boolean compatibility, configuration parsing and boundaries, config-manager rejection, and the resulting classifier request and approval decision.

GitOrigin-RevId: dc018b53b782e2d6b4d6a795cca4945130a71678
2026-08-14 19:11:41 +00:00
iceweasel-oai
d8d7ca73f8 Enable unified exec by default on Windows (#38625)
## What changed

- Enable the stable `unified_exec` feature by default on every platform.
- Update cross-platform integration test expectations so `exec_command` and
  `write_stdin` are exposed on Windows instead of `shell_command`.

GitOrigin-RevId: e5c864bacbde7eab109e7a0e399b8f7b843b384c
2026-08-14 19:07:15 +00:00
Benjamin Carlsson
1426592fcc Open notes when accepting the request input Other option (#38624)
## What changed

When the generated `None of the above` option is selected in a request user
input prompt, route Enter and configured accept actions to the notes editor
instead of submitting the response. Tab continues to open the same editor.

## Testing

Add a regression test confirming that both Tab and Enter reveal the notes editor
without submitting the response.

GitOrigin-RevId: 8a563d81a82a11ca0b26c2b1100053f13be8e8f4
2026-08-14 19:02:27 +00:00
jif
42b5f05cef Preserve MCP namespace descriptions in the tool catalog cache (#38623)
## What changed

Keep MCP namespace descriptions when publishing tool definitions to the
process-scoped catalog cache. Cached definitions now expose the server
instructions to the model before a lazily started MCP connection finishes
initializing.

## Testing

Update the cached MCP startup integration test to verify that cached namespace
descriptions retain the originating server instructions.

GitOrigin-RevId: d1b73381852fd43c6b54a7c22a599474c774dc20
2026-08-14 18:35:13 +00:00
Adam Perry @ OpenAI
478215c5c1 Preserve large gRPC code-mode tool errors (#38621)
## Why

Code-mode tool failure messages larger than 64 KiB were truncated before they
reached the host.

## What changed

- Remove the tool error size limit from the gRPC protocol and host validation.
- Forward failed tool completion messages without truncation.

## Testing

- Verify that a multibyte error larger than 64 KiB is preserved exactly.

GitOrigin-RevId: 264ae4ba4adea5c19b669e4f41ccfd41a0c30fb5
2026-08-14 18:30:45 +00:00
rhan-oai
395723b238 Source multi-agent instructions from the model catalog (#38619)
## What changed

- Add model-catalog messages for root and subagent roles, explicit delegation, and delegation hints.
- Resolve role instructions in config, catalog, then bundled-default order, while preserving empty values as an explicit way to suppress fallback text.
- Refresh catalog-provided role and mode instructions when the model changes, and give full-history forks the selected child model's subagent role without retaining the parent's role guidance.
- Keep existing config overrides and reasoning-effort behavior, including proactive delegation for ultra reasoning effort.

## Testing

- Cover message deserialization and preservation through model overrides.
- Cover precedence, empty overrides, model switches, resumed sessions, and full-history subagent forks.

GitOrigin-RevId: 4625cf7c6a5490176adddfaa0fb99100707daea9
2026-08-14 18:26:20 +00:00
felixxia-oai
14a6813fa8 Apply Guardian policies to v2 risk classification (#38618)
## What changed

- Include the resolved Guardian security policy in the v2 classifier's developer instructions.
- Prefer an explicit `guardian_policy_config`, then the reviewer model's catalog policy, and finally the bundled policy.
- Limit the combined classifier instructions and policy to 10,000 tokens.

## Testing

- Cover configured and catalog policies, policy truncation, and the bundled-policy fallback.

GitOrigin-RevId: 1b84d70081dd924e214636b4b470d57a63d7cb88
2026-08-14 18:21:22 +00:00
Celia Chen
086396f7f6 Prioritize global models in the Bedrock Runtime catalog (#38617)
## What changed

Group Amazon Bedrock Runtime models by routing variant so every Global model
appears before the US cross-region models, while preserving the base model order
within each group. Update catalog priorities and tests to match the new order.

GitOrigin-RevId: e5a55d61163404172a418cb794b25f3b27f83618
2026-08-14 18:01:06 +00:00
jif
fa595fbab8 Route escalated approval retries through Guardian (#38616)
## What changed

- Bypass extension approval contributors when an approval request includes a retry reason, ensuring the retry is reviewed by Guardian.
- Add coverage showing that Guardian can deny an escalated retry even when an extension contributor would approve it, and that the retry reason is included in the review request.

GitOrigin-RevId: c8f556623f26fea605898d4b8e89679cebd63e65
2026-08-14 17:56:38 +00:00
Adam Perry @ OpenAI
0fce933290 Remove the gRPC code-mode enabled tool limit (#38615)
## What changed

Stop rejecting code-mode execute requests solely because `enabled_tools`
contains more than 1,024 definitions. Continue validating each tool definition
while converting the request.

GitOrigin-RevId: da83b61c5f659e6aa221ed663821edcdfc29d3f1
2026-08-14 17:50:01 +00:00
jif
f535950eaa Skip stale Guardian risk score persistence (#38612)
## Why

Concurrent Guardian samples can finish out of order. A stale result rejected by
the thread's latest-score update could still be appended to the rollout.

## What changed

Return without persisting a risk score when `insert_if` rejects it as older than
the score already stored on the thread.

GitOrigin-RevId: 7b2c895742753a19dfe48fd7f8ecd912d3cd3af6
2026-08-14 17:41:34 +00:00
cooper-oai
fb5aa093e0 Support workload identity in remote exec-server auth (#38610)
## Why

Remote exec-server registry requests need to refresh managed credentials before sending a request. Static auth-header resolution cannot perform the asynchronous token exchange required by workload identity.

## What changed

- Add asynchronous auth-header resolution to `AuthProvider`, with the existing static-header behavior as the default.
- Resolve fresh managed credentials for each remote environment registry request while preserving the expected account and workspace identity.
- Load the cloud configuration bundle during remote exec-server startup when workload identity is selected.

## Testing

- Update the managed-auth and environment-registry auth tests to exercise asynchronous header resolution.

GitOrigin-RevId: 5d60f1127467aaacdb5d1a8f3d92278bc4bf2e29
2026-08-14 17:32:37 +00:00
jif
c0d59bf614 Fix the missing resume picker history mode import (#38608)
Import `ThreadHistoryMode` where the resume picker uses it for session
selection state and history mode tracking.

GitOrigin-RevId: ae8b30ad959c872d35eb2a38d514e67ecf1275f1
2026-08-14 17:18:26 +00:00
Adam Perry @ OpenAI
a3cb1c14fb Allow larger gRPC code mode tool descriptions (#38606)
## What changed

Stop rejecting gRPC execute requests when an enabled tool's description exceeds
16 KiB. Identifier, tool-count, and subscription-filter limits remain in place.

GitOrigin-RevId: c30dbf555e691b02bd6d6a835a0c40f671b0d172
2026-08-14 16:42:32 +00:00
Charlie Marsh
3b3a91ccf0 Fix resume transcript preview line selection (#38605)
## What changed

- Scan legacy and paginated transcript items newest-first and stop after collecting the six newest nonblank preview lines.
- Keep paging through bounded paginated history when newer items do not produce visible preview text, while preserving chronological display order and assistant markdown rewrites.
- Move transcript preview loading into a focused module and add coverage for reverse scanning, legacy/paginated parity, and paging past invisible items.

GitOrigin-RevId: 576afe37d034b32db363691bd4f401815fffaea7
2026-08-14 16:37:40 +00:00
Charlie Marsh
1bb6384c19 Avoid paginated resume requests for verified legacy rollouts (#38604)
## Why

Resuming a known legacy rollout with `excludeTurns` first requires the TUI to
retry after the app server rejects paginated history loading.

## What changed

- Carry the history mode reported by the session picker into the resume flow.
- For embedded app-server sessions, resume a legacy rollout with its turns when
  the rollout is still legacy and background migration cannot race the check.
- Keep requesting paginated history when the picker metadata may be stale,
  migration is enabled or active, or the app server is remote.

## Testing

Added coverage for picker server replacement, background migration, maintenance
lock contention, stale history metadata, and paginated resume requests.

GitOrigin-RevId: 2300d1ad93a817eeeeb8f8ed60b073400ddf7106
2026-08-14 16:33:28 +00:00
Dylan Hurd
588ee89324 Isolate Guardian reviewer sessions from parent extensions (#38602)
## What changed

- Start Guardian reviewer delegates with an empty extension registry while preserving extension inheritance for other delegates.
- Disable `GuardianV2` in the reviewer session configuration so reviews do not trigger nested Guardian processing.
- Cover both the reviewer isolation and continued delivery of parent-session tool lifecycle events.

GitOrigin-RevId: 858b3e23d1c06177e443fb0779fcf06cc5294fc2
2026-08-14 16:28:42 +00:00
Adam Perry @ OpenAI
da898490fc Make unbounded connection retries configurable (#38601)
## What changed

- Add the stable, default-enabled `unbounded_connection_retries` feature.
- Require the feature for unbounded sampling retries after connection failures.
  When disabled, connection failures use the existing bounded retry and
  transport fallback path.
- Expose the feature in the generated configuration schema.

GitOrigin-RevId: bd80d02ef065ba924f805c29b639752fc817ed64
2026-08-14 16:23:17 +00:00
jif
3360f4a909 Install Guardian V2 in the app server (#38597)
## What changed

- Add `codex-guardian-v2` to the app server's dependencies.
- Install its feature-gated thread, approval-review, and tool lifecycle contributors with access to authentication and thread management.

GitOrigin-RevId: 25672032ccb541138c8c87c9f27f2a7423b2eee0
2026-08-14 15:59:34 +00:00
jif
742edd6c16 Prioritize new Guardian classifications under load (#38596)
## What changed

- Expand the Guardian sampling pool from 8 to 16 WebSocket connections.
- When the pool is full, supersede the oldest request that has already produced a score before replacing an unfinished classification. Treat superseded classifications as a no-op in the extension.
- Stop response WebSocket work when its event consumer is dropped, including while waiting for the connection lock or draining a completed sample.
- Allow retries across both initially warmed connections for retryable stream failures.

## Testing

- Add a concurrent sampler test that fills the pool and verifies scored drains are replaced before an unfinished classification.

GitOrigin-RevId: 9e4df7b01516094726ff3e11878f8ca742d21b03
2026-08-14 15:55:01 +00:00
jif
23094236ac Let extensions resolve approval reviews before Guardian (#38592)
## What changed

- Give approval review contributors the first opportunity to resolve an action when automatic review is not required for the model.
- Return an extension's decision directly and fall back to Guardian when no extension claims the review.
- Update Guardian V2 to approve actions below its risk threshold through the review contributor API, and remove the extension approval-policy API.

## Testing

- Verify that an extension approval bypasses both the Guardian model call and the user approval prompt.
- Cover Guardian V2 review decisions for low-risk, high-risk, and disabled threads.

GitOrigin-RevId: 5be56941143aff6f358b75b5f8101b7c047147ae
2026-08-14 15:24:10 +00:00
felixxia-oai
5bc8da6d78 Bound Guardian V2 tool actions before risk classification (#38586)
## Why

Oversized tool arguments could make Guardian V2's model-visible action unbounded.

## What changed

- Cap serialized actions at 10,000 tokens while preserving the trusted tool name and `call_id`.
- Truncate nested string values evenly across the available budget.
- Omit lower-priority fields when the JSON structure alone exceeds the limit, while retaining useful action fields when they fit.

## Testing

Add coverage for oversized nested strings, arrays, and object keys, including spoofed identity fields and omission-marker collisions.

GitOrigin-RevId: 13f7dc93da3761db18a34087225196318a3b8098
2026-08-14 14:33:39 +00:00
jif
7c194ff24b Honor cloud-managed requirements in feature listings (#38581)
## What changed

- Load configuration for `codex features list` through the shared cloud-aware
  loader so managed feature requirements are reflected in the reported state.
- Reuse that loader for MCP commands.

## Testing

- Add an integration test showing that a cloud-managed requirement can disable
  `fast_mode` in `codex features list` without rewriting the user's
  `config.toml`.

GitOrigin-RevId: 2bdbc76dfa40ceb29e8293556f4b56cabbd2c43f
2026-08-14 13:50:10 +00:00
jif
2a452d7dc1 Keep the latest Guardian risk score during concurrent sampling (#38580)
## Why

Concurrent Guardian samples can finish out of order, allowing an older sample to
replace the thread's newer in-memory security risk score.

## What changed

- Record when each `SecurityRiskScore` sample starts while preserving compatibility
  with records that have no timestamp.
- Add `ExtensionData::insert_if` so checking and replacing a typed attachment happen
  under the same lock.
- Update Guardian V2 to replace the thread's risk score only when the completed sample
  is newer than the stored one.

## Testing

- Verify concurrent conditional inserts retain the newest value.
- Verify Guardian-generated risk scores include a sampling timestamp.

GitOrigin-RevId: 6543a4e5497da4c623331050789e91613f931c4b
2026-08-14 13:44:51 +00:00
Owen Lin
42d842a91a Report thread storage changes after rollout migration (#38579)
## What changed

Measure thread storage before and after `codex migrate-rollouts --apply` and
include the formatted sizes in the human-readable migration report. The total
includes active and archived rollout files, the thread history database, and
its WAL and shared-memory sidecars.

Keep dry-run and JSON output unchanged, and omit the storage summary if either
measurement fails.

GitOrigin-RevId: 1775c0b6849570a814becd94e196dd05908667a1
2026-08-14 13:40:35 +00:00
jif
1c4f42863c Require automatic review for high-risk Guardian v2 actions (#38569)
## What changed

- Add a thread-scoped extension API for requesting automatic review, resolving
  multiple contributors to the strongest requirement.
- Have Guardian v2 request automatic review when it is enabled and the latest
  `action_risk` score is at least `0.8`.
- Honor extension review requirements across tool approvals, MCP elicitation,
  and remote execution while preserving host configuration and managed-policy
  constraints.

## Testing

- Cover contributor registration and requirement resolution.
- Verify Guardian v2 requests review at the threshold, returns to the default
  below it, and does not affect threads where the extension is disabled.
- Exercise extension-requested strict review with a remote model override.

GitOrigin-RevId: 6329fc23585d5cc7173c39bb9952c3109ec3884b
2026-08-14 12:37:13 +00:00
felixxia-oai
d40dfcc3c7 Run tool start callbacks after pre-tool hooks (#38568)
## What changed

- Invoke `ToolLifecycleContributor::on_tool_start` only after pre-tool hooks
  have finalized the invocation.
- Pass hook-rewritten arguments and the post-hook conversation snapshot to the
  callback.
- Skip the start callback when a hook denies execution or supplies input that
  cannot be applied.

## Testing

Add lifecycle tests covering rewritten input, hook-added context, denied tool
calls, and invalid rewritten input.

GitOrigin-RevId: 936efaf4a8c35321f9982ff2a33bb8bece1ebf39
2026-08-14 12:21:08 +00:00
jif
aa905bb962 Store security risk scores as a snapshot (#38567)
## What changed

- Replace the single `category` and `score` fields in `SecurityRiskScore` with a
  deterministic map of category names to scores.
- Validate the full Guardian V2 classifier response before storing it as one
  thread extension value and, for non-ephemeral threads, one rollout item.
- Cover serialization and rollout loading with snapshots containing multiple
  score categories.

GitOrigin-RevId: 24525de691f96bd34ccfe04acab2f01aca309837
2026-08-14 12:16:51 +00:00
felixxia-oai
08c48e3700 Reuse compatible parent compactions in Guardian V2 (#38564)
## Why

Guardian V2 classification requests should retain compacted conversation context only when the parent and classifier model configurations can share the encrypted compaction.

## What changed

- Resolve the classifier model's compaction compatibility hash when initializing its sampler.
- Add the latest encrypted parent compaction to classification input when both models have the same non-empty compatibility hash.
- Ignore missing, empty, or unidentified compactions, and omit parent context when compatibility cannot be established.

## Testing

- Cover latest-compaction selection and invalid compaction data.
- Cover matching, mismatched, missing, empty, and changed compatibility hashes.

GitOrigin-RevId: f0fcb09fd028cd2f397fdbc3d541b4bc464b0beb
2026-08-14 11:55:10 +00:00
jif
3134ab6572 Restrict filesystem helper sandbox access (#38561)
## Why

Filesystem helpers only need access to their own executables. Granting access to
the containing directories also makes unrelated sibling files readable.

## What changed

- Allow filesystem helpers to read the Codex and Linux sandbox executables
  directly instead of their parent directories.
- Give macOS filesystem helpers a narrower Seatbelt profile that omits the
  normal process sandbox's `/Applications` read access.
- Preserve the existing platform defaults for normal sandboxed processes.

## Testing

Add unit and integration coverage for direct and symlinked sibling-file access,
sandboxed copies, allowed workspace reads, and the macOS-specific Seatbelt
profiles.

GitOrigin-RevId: 9380c6868a695bdf2275baa74bd5e8a30a64ba1b
2026-08-14 11:47:12 +00:00
felixxia-oai
486df09a00 Improve Guardian transcript context selection (#38558)
## What changed

- Replace the single byte limit with token-derived per-entry and separate message and tool transcript budgets.
- Preserve the first and latest user messages, then fill the remaining budgets with recent user, assistant, and tool context while limiting retained non-user entries.
- Truncate oversized entries in the middle without splitting UTF-8 characters, and report the approximate omitted token count.
- Exclude reasoning from the default transcript sources while retaining support when it is explicitly configured.

## Testing

- Cover oversized message and tool-result truncation, user-message retention, separate tool budgeting, configured reasoning, and the resulting Guardian sampling request.

GitOrigin-RevId: aaeaa5fb5d902b5a1581a8a819bf34724bde794b
2026-08-14 11:17:50 +00:00
felixxia-oai
db5b4a4cfb Preserve Guardian transcript boundaries in sampling input (#38555)
## What changed

- Represent Guardian sampling input as an ordered list of text entries.
- Send each transcript entry and approval-request delimiter as a separate input
  content item while preserving the existing transcript byte limit.

## Testing

- Update transcript and sampler tests to verify entry boundaries, multi-item
  request serialization, and bounded transcript behavior.

GitOrigin-RevId: 433e2e6ca18ce9288025715f5956295bfe852e98
2026-08-14 11:02:01 +00:00
jif
a44f266a4a Initialize Guardian V2 samplers per thread (#38553)
## What changed

- Register Guardian V2 for thread startup as well as tool-call lifecycle events.
- When `guardianv2` is enabled, create and store a thread-local Luna sampler using the thread's model provider, authentication policy, session metadata, originator, and service tier.
- Emit a warning if sampler initialization fails, and skip tool classification when the thread has no sampler.

## Testing

- Update the extension test to enable `guardianv2`, run thread startup, and verify tool-call sampling through the initialized connection.

GitOrigin-RevId: 5ab732e85c9827495c67f57117eb5be450a7f90e
2026-08-14 10:58:56 +00:00
jif
8630bb3cae Record Guardian V2 risk scores on threads (#38540)
## What changed

- Parse and validate the structured risk scores returned for each classified tool call.
- Publish each score through thread extension data and persist it in rollout history for non-ephemeral threads.
- Report parsing, thread lookup, validation, and persistence failures through the extension warning sink.

## Testing

- Extend the Guardian V2 tool-classification test to verify that the sampled `action_risk` score is available from thread extension data and written to rollout history.

GitOrigin-RevId: 46614fe0d686736dfb1f09398c6fcd848c7d8b2c
2026-08-14 09:27:41 +00:00
Dylan Hurd
4eff3b788b Configure Guardian sampling for responses lite (#38527)
## What changed

- Opt Guardian V2 Luna sampler WebSocket connections into responses lite.
- Send `reasoning.context` as `all_turns` on sampling requests.

## Testing

- Extend sampler and extension tests to verify the connection header and reasoning context.

GitOrigin-RevId: 4b0263c01b8259ed5268c222dcf7559912baba8c
2026-08-14 08:16:23 +00:00
sayan-oai
fdbab67c66 Carry environment config in turn selections (#38521)
## What changed

- Add `EnvironmentConfigState` to `TurnEnvironmentSelection` so each environment attachment can inherit thread configuration or carry ready configuration for its login-shell policy and selected capability roots.
- Apply ready configuration during thread creation and thread-settings updates while reusing the existing environment connection.
- Validate selection configuration before applying settings, rejecting unsupported pending state and invalid capability roots.

## Testing

- Cover attachment-scoped capability roots, ready configuration at thread startup and turn updates, and rejection of invalid selections without changing existing settings.

GitOrigin-RevId: 86cf8f2148fdaa746ea0dd12890ab7b1577d7587
2026-08-14 07:52:38 +00:00
rka-oai
86b1123ff6 Enable parallel tool calls for all model prompts (#38499)
## What changed

- Set `parallel_tool_calls` for regular and remote compaction prompts without consulting model metadata.
- Remove `supports_parallel_tool_calls` from `ModelInfo`, the bundled model catalog, and related fixtures.
- Preserve the existing Responses Lite behavior that disables parallel tool calls at request construction.

GitOrigin-RevId: 49552bdf97e71fa57325abb49f8adc0c88f1401a
2026-08-14 05:49:33 +00:00
Adam Perry @ OpenAI
636e505c5c Verify bundled bwrap in Bazel builds (#38494)
## Why

Bazel builds did not embed the SHA-256 digest used to verify the bundled
`bwrap` binary at launch.

## What changed

- Strip the Bazel-built `bwrap`, generate its SHA-256 digest, and pass that
  digest to `codex-linux-sandbox` through a compiler environment file.
- Report digest verification failures with exit code `8` instead of panicking.
- Add a Bazel integration test that accepts the original bundled binary and
  rejects a tampered copy.

GitOrigin-RevId: 224eeb6ae3d1bfb100e048b296f75fd8ba109581
2026-08-14 05:03:39 +00:00
Matthew Zeng
45c9c74e29 Reuse pending MCP startups during reconciliation (#38493)
## Why

MCP runtime reconciliation can occur while an unchanged server is still
starting, such as when capability roots become available. Reconciliation
should neither wait for that startup nor initialize the same server again.

## What changed

- Reuse an active, uncancelled pending MCP connection when its identity,
  catalog limit, and protocol mode still match.
- Publish the shared startup result through the reconciled server view while
  applying its updated tool filter.
- Preserve cancellation when a reused pending server is subsequently removed.

## Testing

Added coverage for filtered-tool publication, disabling a reused pending
server, and capability-root reconciliation during Codex Apps startup.

GitOrigin-RevId: 690cf8695e75b341d097d7d22b5da6993b24e787
2026-08-14 04:49:14 +00:00
Dylan Hurd
4e5a08feb6 Enforce strict auto-review for MCP tool calls (#38492)
## What changed

- Route MCP tool calls through the automatic reviewer when strict auto-review is enabled, even when the approval policy, tool annotations, or a remembered session decision would otherwise skip review.
- Pass the strict auto-review flag into the MCP approval request so reviewer selection follows the turn setting.
- Update MCP approval and turn-metadata tests to cover the forced review path and confirm that it does not request user input.

GitOrigin-RevId: 2c0b5f4dc1a15cb2fb827e4b21e69167fdcf3e56
2026-08-14 04:46:04 +00:00
dozaki-openai
cbe85e117b Search selected plugin apps before falling back (#38484)
## What changed

- When an explicitly selected plugin has apps available, instruct the model to use `tool_search` to discover relevant tools before falling back to unrelated or built-in tools.
- Scope the guidance to the request that selected the plugin and omit it for plugins without available apps.

## Testing

- Cover app-specific rendering, non-app plugins, bounded instruction output, and app enablement in dual-surface plugin requests.

GitOrigin-RevId: 1b9501cc403d2817334a9307e683e99c140a794a
2026-08-14 02:54:38 +00:00
pmccrary-oai
9d012ca4f5 Include agent names in turn metadata (#38483)
## What changed

- Add the canonical agent path as `agent_name` in Responses API turn metadata.
- Fall back to `/root` when a session has no canonical agent path.
- Reserve `agent_name` from client metadata overrides and omit it from MCP request metadata.

## Testing

- Cover root, named subagent, legacy subagent, MCP, and client metadata merge cases.

GitOrigin-RevId: a9adcc079a51beae7d621aba24bcd759ceaa408e
2026-08-14 02:32:05 +00:00
felixxia-oai
5620bab61c Add bounded skill model delegation instructions (#38475)
## What changed

- Add `SkillModelDelegationInstruction` for skills that request Luna while running on Sol or Terra.
- Resolve Luna only when it is available in the current provider namespace.
- Bound and validate model identifiers, skill names, and the rendered instruction before exposing it to callers.

## Testing

- Cover supported parent models, provider namespace matching, unavailable or unsafe targets, instruction rendering, and size limits.

GitOrigin-RevId: 188a88c36ca32689cbf1e53465283dff4137cb15
2026-08-14 01:49:00 +00:00
Tamir Duberstein
5c6f498b0e Stop generating accepted-line fingerprints (#38473)
## What changed

- Replace accepted-line fingerprint generation with diff-based added and deleted line counts.
- Keep `line_fingerprints` in the analytics payload as a statically empty field for schema compatibility.
- Remove the fingerprint types and public parsing API that are no longer used.

## Testing

- Update diff parsing tests to cover accepted-line counts without generating fingerprints.

GitOrigin-RevId: 4247c76c7b56b1a323f70f960823937352b84f7b
2026-08-14 01:44:33 +00:00
Celia Chen
d5e256ceb2 Add an Amazon Bedrock Runtime provider (#38470)
## What changed

- Add the built-in `amazon-bedrock-runtime` provider for regional
  `bedrock-runtime` OpenAI-compatible endpoints.
- Use endpoint-specific SigV4 service configuration while preserving bearer
  token auth and per-provider AWS profile, region, and transport overrides.
- Provide global and US cross-region GPT-5.6 model variants, prefer global
  routing for fallback and background tasks, and disable unsupported web search.

## Testing

- Cover provider configuration and capabilities, endpoint and auth selection,
  runtime catalog ordering, model fallback, and web search omission.

GitOrigin-RevId: d158ab74a5768e1b4c61063365a0c06735313322
2026-08-14 01:23:59 +00:00
felixxia-oai
3711943d11 Parse model annotations from skill frontmatter (#38467)
## What changed

- Add an optional `model` field to parsed and loaded skill metadata.
- Recognize `model: luna` while ignoring unsupported model values without
  preventing the rest of the skill metadata from loading.

## Testing

- Cover supported, absent, unsupported, and repaired frontmatter model values.

GitOrigin-RevId: 376b65555e81d29a5eae93cc4c2633aacf7cd7c8
2026-08-14 00:54:45 +00:00
Owen Lin
c6dee5f49f Preserve thread subscriptions across revert reloads (#38463)
## Why

The connection that requested `thread/revert` may close while the thread is
being reloaded, but the replacement listener still needs to serve the thread's
existing subscriptions.

## What changed

Restart the listener task from the preserved thread state after a paginated
thread reload instead of attaching it through the requesting connection.

GitOrigin-RevId: 603ec3331c59bdde48faa2f4cb0fd1032ec8dbd8
2026-08-14 00:27:39 +00:00