Commit Graph
257 Commits
Author SHA1 Message Date
Andrei Eternal c8a8295e79 Attribute turn metrics to the models used during the turn (#44656)
## Why

Turn completion metrics used session-level telemetry, which could label usage with the wrong model after a model switch. A turn can also include compaction and responses from different models, so a single token total cannot attribute usage correctly.

## What changed

- Accumulate token usage by model using each response's resolved step settings, and emit one histogram sample per model and token type per turn.
- Preserve zero-valued samples when no usage is reported, without adding a sample for an unused selected model when only compaction ran.
- Use turn telemetry for tool-call, memory, network-proxy, and running-process metrics.

## Testing

Add regression tests for model switches, step-level model changes, compaction followed by a response or an aborted turn, and turns without responses. Verify repeated requests aggregate into one sample per model and earlier session usage is excluded.

GitOrigin-RevId: 03ab1dcde7da0b16b3a500a9a05e4c86583ee585
2026-09-10 21:58:58 +00:00
Matthew Zeng 8570091e14 Honor thread-level plugin exclusions across runtime capabilities (#44655)
## What changed

- Apply `disabled_plugin_ids` to plugin skills, recommendations, hooks, and MCP servers without modifying shared plugin state.
- Activate selection changes when the next task starts, keeping pending settings from changing the active runtime.
- Hide app tools contributed only by disabled plugins from model tool catalogs and search. Preserve connectors shared with enabled plugins and existing direct Apps RPC behavior.
- Include plugin identity in MCP approval keys and use that identity when persisting plugin tool approvals.

## Testing

Add regression coverage for disabling and re-enabling plugin skills and tools, deferred activation of settings, shared connector ownership, selected plugin servers, executor hooks, and plugin-specific approval keys and persistence.

GitOrigin-RevId: 1153ff6aa7ab6953d50e838da05203904ecd2956
2026-09-10 21:51:34 +00:00
felixxia-oai 287e4f7dbf Preserve Guardian authorization evidence until request budgeting (#44570)
## Why

Transcript limits could shorten user instructions or omit later messages before Guardian evaluated the complete request budget, losing restrictions or prior approvals even when they would fit.

## What changed

- Keep user messages and manual approvals complete and in source order through transcript collection and retention.
- Preserve historical instructions while synchronous review can still compact history. If the final request still cannot fit, discard optional evidence before shortening older historical entries with truncation markers, preserving later restrictions where possible.
- Warn that shortened instructions and approvals are incomplete and that missing evidence does not authorize actions.
- Let asynchronous review defer to synchronous review when complete instructions exceed its budget.

## Testing

Add coverage for complete instruction retention, approval and restriction ordering, marked truncation of oversized Unicode text, compaction before truncation, and asynchronous fallback when instructions do not fit.

GitOrigin-RevId: e6159e876231347986b937ee7e07f944b83bfbe6
2026-09-10 16:02:36 +00:00
felixxia-oai 9c4879f3a5 Preserve complete actions in Guardian approval reviews (#44569)
## Why

Truncating action arguments can leave approval reviewers evaluating incomplete actions. Large actions need complete review input and explicit handling when they exceed the review budget.

## What changed

- Remove action truncation and the fixed synchronous action byte limit. Admit complete actions against the whole-request budget, splitting long text losslessly into bounded transport parts and accounting for their framing.
- Route actions exceeding the asynchronous action budget to synchronous review. Prevent cached scores from covering oversized calls, including expanded approval arguments, while allowing later small actions to recover score reuse.
- Request user approval when optional review exhausts its local input budget. Keep required review and compaction service failures closed to approval, and retire exhausted review sessions.

## Testing

Add coverage for complete large-action delivery, optional user fallback, required-review denial, subsequent review recovery, async overflow through MCP approval routing, and lossless text splitting with budget accounting.

GitOrigin-RevId: 08f06e68a94b2f779480dd6b6cf5bc30b241f6dd
2026-09-10 16:01:02 +00:00
jif ee93abb690 Preserve incoming prompts when pre-turn compaction fails (#44487)
## Why

Pre-turn compaction runs before incoming input is recorded, so failures could leave an accepted prompt out of conversation history. Reporting the error before prompt hooks finish also lets clients steer follow-up input into a turn still preserving its prompt.

## What changed

Record input and run prompt hooks on every pre-turn compaction failure. Defer local and remote compaction error events to `run_turn`, after prompt preservation completes, while retaining the remote error prefix.

## Testing

Extend the remote compaction output-validation test to trigger automatic compaction and assert that the prompt is emitted before the error, saved exactly once, and followed by no additional model request. Retain the assertion that usage is recorded for invalid compaction output.

GitOrigin-RevId: ab7ccb259836c084173e0e81dbbd4c610e95a6c8
2026-09-10 10:08:11 +00:00
felixxia-oai fcd90d8f07 Enforce complete request budgets for Guardian reviews (#44281)
## Why

Review evidence can fit its own limits while exceeding the reviewer context window once history, tools, output format, and reminders are included. Tool continuations can also push an existing review over budget.

## What changed

- Select evidence after tools and turn context are resolved, reserving request overhead and a 256-token margin. Omit optional evidence while preserving required sections, and reject required evidence that cannot fit.
- Keep incoming evidence pending until budget selection succeeds so cancellation and compaction cannot record unselected input in reusable history.
- Check the assembled prompt before sampling, including retries and tool continuations. Retry budget failures with summary compaction once per model step, and retire reviewers whose budgets remain exhausted. Avoid history-resetting token-budget compaction for this recovery.
- Record section costs after budget selection.

## Testing

Add unit and integration coverage for required-evidence rejection, optional-evidence omission, reminder overhead, cancellation safety, incoming-review and tool-continuation compaction, and reviewer retirement after ineffective compaction.

GitOrigin-RevId: 2f7ef191b3f46032d295ad363eb7debf9d7a9684
2026-09-09 19:21:08 +00:00
rka-oai 3dc1e2a584 Always use streamed remote compaction for supported providers (#44255)
## What changed

- Route manual and automatic remote compaction through the v2 implementation.
- Retire the `remote_compaction_v2` feature toggle while continuing to advertise it in the model client beta features header.
- Migrate compaction tests from `/responses/compact` to streamed `/responses` requests with `compaction_trigger` items and remove legacy parity coverage.

## Testing

Add streamed compaction coverage for tool search output trimming and session base instruction token estimates. Update app-server tests to verify started/completed notifications for both local and streamed remote compaction.

GitOrigin-RevId: cea392a6166a9b116ff83708fd9126d5546476e4
2026-09-09 18:04:04 +00:00
Abhinav b64de2f3ad Use the originating model when recording conversation history (#44243)
## Why

History recording used the model captured at turn start, so switching models within a turn could apply stale image preparation and tool-output truncation rules.

## What changed

Pass explicit model metadata through conversation recording. Record model responses and tool results using their originating step's settings, and capture the current model for pending input recorded during startup interruption or task completion.

## Testing

Add regression coverage for tool results across a model switch, explicit model selection for plain and annotated history, and image input retained during interruption and completion. Extend model-switching tests to verify that prepared media and item identity survive switching back to a multimodal model in live, resumed, and forked threads.

GitOrigin-RevId: 5bc750c30c334acb9b802cb80fdea01daffa84a1
2026-09-09 17:46:00 +00:00
ningyi-oai 0d46c252b3 Encapsulate executed tool call metadata recording (#44002)
## What changed

Introduce `ExecutedToolCalls` to own optional shared recorder state and centralize feature checks, call recording, and accepted result source lookup. Update session, tool runtime, MCP, and Code Mode callers to use this interface.

Move prompt attachment, retry handling, and metadata budgeting into the `request_metadata` module, preserving existing recording and request behavior.

## Testing

Add coverage for all session and turn feature combinations, lazy result source lookup, and unchanged prompts when recording is disabled. Retain the existing metadata budget and cell completeness tests in the new module.

GitOrigin-RevId: bc19f34317232b6599c7ec71c15db4a80123ff6f
2026-09-09 02:33:17 +00:00
felixxia-oai 35d9e4bc4d Preserve reasoning effort through compaction and reset it on success (#43796)
## Why

With reasoning effort overrides enabled, compaction used the selected effort even when sampling requests still used an earlier pinned effort. Successful compaction also left that old pin active in the new context window.

## What changed

- Use the pinned request effort for local and remote compaction when it matches the model.
- Resolve fallback models' effort without mutating the live pin, preserving it if compaction fails.
- Reset the pin after successful compaction so the next sampling request establishes the selected effort as its baseline without a redundant `configuration_update`.

## Testing

Add regression coverage for remote compaction success and failure, fallback model effort selection, pin preservation during fallback lookup, and compaction after resuming with overrides disabled.

GitOrigin-RevId: 35e666b0c8026553500715ddbe7bef02cd759f4a
2026-09-08 10:46:07 +00:00
felixxia-oai 31ccaf40c2 Pin request reasoning effort while configuration overrides are active (#43795)
## Why

Reasoning-effort changes should preserve the request baseline while trusted `configuration_update` items carry the selected effort.

## What changed

- Pin request-level reasoning effort for the current model when reasoning-effort overrides are available, including during WebSocket prewarm.
- Re-establish the selected effort after rollback, resume, or a model switch, refreshing the trusted override even when the selection is unchanged.
- Continue using the selected request effort when overrides are unavailable.

## Testing

Add regression coverage for WebSocket prewarm and reconnect, rollback, resume, and model switches. Update transition tests to assert that request-level effort stays pinned while configuration updates change.

GitOrigin-RevId: 89cfee6289543230347b0488d1388190b4da7216
2026-09-08 10:41:15 +00:00
Benjamin Carlsson b01c3986fd Add live WebRTC voice conversations to the TUI (#43581)
## What changed

- Add feature-gated `/voice`, `/voice mute`, and `/voice stop` commands with local WebRTC audio and app-server signaling.
- Show live transcripts, conversation status, and microphone and speaker levels. Preserve captions and undelivered answers across thread switches.
- Speak final answers from voice handoffs while keeping delegated reasoning and commentary hidden and typed answers unspoken.
- Retry eligible startup failures once and clean up voice sessions on thread switches and disconnects.
- Stop voice and block late handoffs after a misalignment policy violation. Remove realtime event payloads and spoken text from receipt and debug logs.

## Testing

Add coverage for voice command mapping, rejected starts, disconnect cleanup, delegated final-answer speech, late handoff rejection, and transcript and footer rendering.

GitOrigin-RevId: 11def1f86b0e023b21d0d92072d6636ccad89815
2026-09-07 20:25:30 +00:00
riley-oai 555b82afa9 Add opt-in MCP user-verification transport (#43352)
## Why

Device-authenticated verification needs an app response outside automated approval and review. Previously, MCP user-verification requests were always cancelled.

## What changed

- Add typed `openai/userVerification` elicitations carrying a title, description, and challenge through core and app-server.
- Route verification to one connection enabled by trusted host activation. Restrict responses to that owner, exclude requests from replay, and cancel on disconnect or authentication changes, including account switches away and back.
- Return proofs in `content`, discard response metadata, and cancel malformed acceptances. Suppress verification response logging and analytics, and send only a generic verification notice to realtime.
- Keep the mode experimental and exclude it from stable schema exports. Capability advertisement remains disabled, and the TUI cancels verification requests.

## Testing

Add coverage for owner isolation, disconnect and authentication races, proof response handling, automated-review bypass, inactive-host cancellation, stable schema filtering, and realtime payload exclusion. Update the MCP integration test to verify that an activated request holds the tool until the client responds.

GitOrigin-RevId: c8c7b6691a3f7756682bbcca9def1df0d1feb906
2026-09-07 04:33:22 +00:00
felixxia-oai 56a8470aa0 Record reasoning effort changes in conversation history behind a flag (#43110)
## What changed

Add the disabled-by-default `reasoning_effort_override` feature for OpenAI models with `use_responses_lite` enabled. Append a trusted `configuration_update` after accepted input when no effort is established in surviving history or the resolved effort changes. Preserve the existing history prefix and continue sending the request-level reasoning effort.

Share effort normalization through `ModelInfo::resolve_reasoning_effort` so requests and history updates resolve `ultra` consistently and translate `persistent` to `disabled`. Exclude other custom effort values from history updates, and compare only against harness-authored configuration items.

## Testing

Add integration coverage for effort transitions, deduplication, history prefix and cache-key preservation, alias normalization, and feature/provider/model gating. Add unit coverage for model-specific `ultra` resolution, fallbacks, and `persistent` translation.

GitOrigin-RevId: c3fe7050058076454d8ea20054a618bca9d48c7c
2026-09-05 21:56:43 +00:00
jif dee21ec1bc Replace Guardian tickets with parent response IDs (#43002)
## What changed

Capture the ID from `response.created` and pass it as `parent_response_id` in Guardian review and classifier request metadata. Clear the current response ID before sampling retries so tool reviews cannot inherit a failed response's ID.

Request Guardian credits with `guardian_credits_requested` on eligible parent requests. Remove the ticket protocol, transport attachment, and ticket-specific body redaction. Keep ordinary app-server client `parent_response_id` values inside `x-codex-turn-metadata` rather than promoting them to Guardian request metadata.

## Testing

Update coverage for HTTP fallback, WebSocket review reuse, missing response IDs, classifier transport retries, and code-mode yield/resume attribution. Extend app-server tests to verify client metadata placement over HTTP and WebSocket.

GitOrigin-RevId: c1665e56382ecec8d260fc2ca0ecd5e20b0b2391
2026-09-05 13:02:49 +00:00
jif cc4b8bdeb8 Propagate response tickets to Guardian reviews (#42758)
## What changed

- Request and parse server-issued Guardian tickets for eligible Codex backend responses.
- Scope each ticket to the response that produced a Guardian review or classifier request, including HTTP and WebSocket retries, without adding it to model-visible or persisted context.
- Validate ticket shape, reserve its metadata keys, redact it from debug and request-body logging, and attach it only at the Guardian transport boundary.

## Testing

- Cover ticket parsing, endpoint and authentication routing, retry scoping, classifier retries, and redaction for compressed and uncompressed requests.

GitOrigin-RevId: 6219dd15c9f19ed3387de516aac1b8e7c9cdb0e3
2026-09-04 12:56:26 +00:00
jif 4e48cd02da Honor model-provided Guardian review policies (#42744)
## What changed

- Add optional per-model Guardian modes for computer use, shell, code mode,
  file changes, MCP, network, and permission requests.
- Apply disabled, synchronous, or adaptive review by action category while
  preserving legacy configuration when model policy is absent.
- Bind cached risk scores to the active model policy, thread settings, and
  environment selections so outdated scores cannot approve later actions.
- Keep unknown review modes on the synchronous path and ignore unknown policy
  fields for forward compatibility.

## Testing

- Cover policy serialization, legacy fallback, scoped tool classification,
  disabled categories, and score invalidation after policy changes.

GitOrigin-RevId: f334a5f4fffb198da66cde3a400233e0df650516
2026-09-04 11:59:18 +00:00
Matthew Zeng 460b63e5f4 Honor explicit plugin mentions during MCP startup (#42406)
## Why

Selected plugin MCP servers may be skipped after the optional startup grace
period. When a user explicitly mentions a plugin or MCP server, the turn should
continue waiting so the requested tools are available to the model.

## What changed

- Track explicitly mentioned plugin IDs and MCP servers across later samples,
  steered input, and same-turn restarts.
- Require matching selected-plugin MCP servers to finish startup before
  capturing the tool catalog, while leaving unmentioned servers optional.
- Match `plugin://` identities independently of display names and ignore
  targeting query parameters such as `app` and `browserFamily`.

## Testing

Added coverage for linked and structured plugin mentions, direct and batched
capability discovery, delayed environment attachment, steering, and same-turn
restarts.

GitOrigin-RevId: c509ef8df45df1a7edb0da6afdb2682dc53a4de8
2026-09-03 01:33:32 +00:00
felixxia-oai 0d502a4230 Support durable reasoning configuration updates (#42328)
## Why

Reasoning configuration changes need to retain their position and trusted provenance when model history is persisted and replayed. Client-injected history must not be able to forge these controls.

## What changed

- Add a typed `configuration_update` response item carrying reasoning effort, including custom model-defined values.
- Persist harness-authored updates with provenance and preserve them across history reconstruction, thread resume, raw response notifications, and agent forks.
- Exclude untrusted configuration updates from model history, strip client-supplied provenance metadata, and reject configuration updates supplied as turn input.
- Export the new item through the JSON and TypeScript app-server schemas and classify it in telemetry and persistence metrics.

## Testing

- Cover serialization, provenance persistence, history filtering and rollback, resume reconstruction, and injection attempts before and after restart.

GitOrigin-RevId: eb5559d2b52b7a931621e7c9812f009ff9fb8939
2026-09-02 18:52:44 +00:00
rka-oai 2c79ee6dac Add structured asynchronous user input requests (#42178)
## What changed

- Replace `send_user_message_async` with `request_user_input_async`, accepting one or more questions with optional suggested answers while allowing the turn to continue.
- Attach structured question metadata to asynchronous agent messages while retaining readable fallback text, and preserve it through app-server events, thread history, and generated schemas.
- Continue enabling the new tool for model catalogs that advertise either the old or new tool name.

## Testing

- Cover tool registration, question validation, emitted items, continued turn execution, history preservation, and backwards-compatible deserialization.

GitOrigin-RevId: ffc49aebde2e854c3c50a122aa52805fec3fc6b3
2026-09-02 02:09:35 +00:00
jif 86b7d12742 Add Guardian V2 analytics events (#42144)
## What changed

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

## Testing

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

GitOrigin-RevId: 0805975f8e8f9d954aa1058ea8c60545a090ae74
2026-09-01 22:02:42 +00:00
jif 612e6491d5 Prewarm shell snapshots for eligible turns (#42137)
## Why

Shell Snapshot V2 previously captured the login-shell environment lazily when a
command started, adding that work to the command path.

## What changed

- Start one asynchronous snapshot capture after turn hooks accept the turn, so
  eligible local `exec_command` calls can reuse it without blocking the model.
- Use the turn's resolved tools, permissions, sandbox, environment, and shell
  configuration, and skip prewarming for unsupported or network-managed cases.
- Cancel outstanding prewarming during shutdown. Keep failed speculative
  captures retryable by real commands without consuming their retry budget.

## Testing

Added coverage for non-blocking capture and reuse, failure recovery, sandbox and
permission isolation, shutdown cancellation, hook-stopped turns, eligibility
gates, and concurrent capture retries.

GitOrigin-RevId: ff24a4f0328c06653e09d03fd0a3db1eccd65030
2026-09-01 21:16:12 +00:00
Owen Lin 5f79a92e39 Persist response token usage in rollout history (#41912)
## Why

Resumed threads need to continue per-turn and per-thread token totals without scanning arbitrarily far beyond the latest compaction checkpoint.

## What changed

- Add durable `TokenUsageRecord` rollout items with response, turn, thread, session, and root-turn attribution.
- Restore accumulated usage on resume and snapshot the latest record plus the compaction response ID in compaction checkpoints.
- Preserve root-turn lineage in persisted turn context, while ensuring forked child threads start with their own usage totals.

## Testing

- Cover usage accumulation across multiple responses and resumed turns.
- Cover local and remote compaction checkpoints, invalid remote compaction output, rollout reconstruction, and fork isolation.

GitOrigin-RevId: ef9e0c4a9102a08a2c382be4cdac68c84353c90a
2026-08-31 21:15:38 +00:00
rka-oai a9519cbcdd Make the update_plan tool opt-in (#41744)
## What changed

- Default `tools.update_plan.enabled` to `false`; users can explicitly enable it to expose `update_plan`.
- Remove bundled `update_plan` guidance from model, collaboration-mode, multi-agent, compaction, prewarm, and goal-continuation prompts when the tool is disabled.
- Preserve custom base instructions, model catalog instructions, collaboration policies, and user goal text even when they mention planning or `update_plan`.

## Testing

- Cover default and explicitly enabled tool registration, prompt consistency across request paths, and preservation of custom instructions.

GitOrigin-RevId: a53964e6e72f98e2557dd3090fdb7caccb956527
2026-08-31 00:53:30 +00:00
rhan-oai 3c062df036 Source async user message descriptions from the model catalog (#41461)
## What changed

- Add model message metadata for built-in tool descriptions.
- Use the active step model's catalog description for `send_user_message_async`, including after a mid-turn model change.
- Fall back to the built-in description when catalog metadata is absent, while preserving an explicitly empty description.

## Testing

- Cover catalog serialization, fallback and empty-value behavior, and model changes within a turn.

GitOrigin-RevId: 72b953214ab9708931065321debf12133f8d2d40
2026-08-29 03:44:01 +00:00
Krish Chainani 0d22692962 Retain the last selected step context for each turn (#41429)
## What changed

- Store the most recently captured execution `StepContext` in the active turn state.
- Keep speculative model-fallback captures from replacing that context, and update it only when the fallback is selected for remote compaction.
- Preserve the retained context when a turn completes or is interrupted.

## Testing

- Add coverage for completed and interrupted turns, successful primary compaction, fallback compaction, and interruption while a fallback is running.

GitOrigin-RevId: c7dee36ec8e88ba2f4fb4a9c0322e5c86a0caf0e
2026-08-28 22:26:23 +00:00
pakrym-oai f742dabc6f Support per-tool MCP output limits (#41421)
## What changed

- Add a positive `output_token_limit` setting to each entry under an MCP server's `tools` configuration.
- Apply the most restrictive limit when plugin and user policies overlap, while keeping approval policy independent.
- Carry the effective MCP output budget in conversation history so tool output, post-tool hook responses, and resumed sessions use the same truncation limit.

## Testing

- Cover configuration parsing, serialization, schema validation, and plugin policy merging.
- Cover MCP output below and above the configured limit, post-tool hook responses, and session resume.

GitOrigin-RevId: d0beb4fca9ba6055d9e1d31c137373b465d50d61
2026-08-28 21:40:23 +00:00
alexsong-oai 41d3dc56a0 Surface model provider authentication recovery progress (#41239)
## What changed

- Emit turn-scoped authentication recovery started and completed events when a model provider refreshes expired credentials.
- Add stable `modelProvider/authRecoveryStarted` and `modelProvider/authRecoveryCompleted` app-server notifications with the thread, turn, provider, and user-facing message.
- Show recovery progress in the TUI and `codex exec`, including Amazon Bedrock session reauthentication.

## Testing

- Cover provider recovery success and failure events, app-server routing, client rendering, and Amazon Bedrock credential refresh.

GitOrigin-RevId: 3010c38d0676f18bced27761cf86dd38344d09f9
2026-08-28 01:14:02 +00:00
Abhinav 0182ff3480 Finalize model-specific tool plans in ToolRouter (#41195)
## Why

Planning tools for a candidate model must not change the selected model or its
Responses Lite tool inventory. Otherwise, preparing a fallback can overwrite
metadata before the current model's request is sent.

## What changed

- Build tool registries and visible specs from an explicit `ModelInfo`.
- Store the finalized tool mode, code-mode names, namespace inventory, and
  terminal and child-management capabilities on `ToolRouter`.
- Publish namespace inventory only after selecting the step's router, and use
  that router's mode when dispatching and recording tool calls.

## Testing

Added coverage for candidate-model isolation, fallback compaction metadata,
code-mode worker selection, normalized tool names, and router capabilities.

GitOrigin-RevId: 56df2de8f42c7e93a4c4dcd117720e40a0cca3db
2026-08-27 20:01:39 +00:00
Dylan Hurd 2c4a95736b Expose response usage metadata in completion events (#41087)
## What changed

- Parse optional `usage_metadata.amount` values from Responses API completion events without converting their string representation.
- Propagate the metadata through SSE and WebSocket streams, regular turns, and remote compaction.
- Add `usageMetadata` to the app-server `rawResponse/completed` notification and its generated schemas.

## Testing

- Cover missing, null, zero, and high-precision amounts, including completion events emitted during compaction.

GitOrigin-RevId: b16d81e8350eab0dc7ff613d0dc9dbed73041713
2026-08-27 07:50:05 +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
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
rka-oai b9c4b9a0cf Support standalone function outputs in turn routing (#40991)
## What changed

- Allow `start_or_steer_turn` to accept named `function_call_output` items without a `call_id`, starting a turn when idle or steering the active turn.
- Preserve standalone outputs as pending input, assign response item IDs, and record them in conversation history without treating them as user prompts.

## Testing

- Cover starting a turn with a standalone output and steering an active turn with one.
- Verify passive response items do not signal pending steering activity.

GitOrigin-RevId: e32b292106b3c5a5f03577b8ebf67d2d7dee4e5d
2026-08-26 22:48:37 +00:00
Francis Chalissery 7276d67081 Expose resumable misalignment details through app server (#40952)
## Why

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

## What changed

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

## Testing

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

GitOrigin-RevId: 329258a444c2cd91d0c57ab1720830b33ddcfac5
2026-08-26 18:42:44 +00:00
faizan-oai 94967e03e5 Add per-turn cyber access program selection (#40687)
## What changed

- Add the experimental `turn/start.cyberAccessProgram` option with `standard`,
  `daybreakBlue`, and `daybreakRed` values.
- Forward the selection as `access_programs.cyber` on Responses, WebSocket,
  and remote-compaction requests made with ChatGPT authentication, while
  omitting it for API-key and custom-provider requests.
- Preserve the per-turn selection across recovery, compaction, and child-agent
  turns without making it a persistent thread setting.

## Testing

- Cover app-server forwarding, authentication boundaries, WebSocket reuse,
  turn recovery, compaction, and child-agent inheritance.

GitOrigin-RevId: d2eb468f365b2214c5099bc21741cf30f8bd2eb5
2026-08-25 19:54:52 +00:00
joeflorencio-openai 68301fa45f Snapshot resolved settings for each model step (#40651)
## Why

Thread settings can change while a turn is running, and delegated review or
compaction steps can select a different model. Each request needs a consistent
set of settings resolved against the model that will execute it.

## What changed

- Capture immutable resolved settings for each model step, including model
  metadata, reasoning options, service tier, approvals, and personality.
- Keep in-flight steps on their captured settings while applying thread updates
  to subsequent work.
- Resolve inherited reasoning summaries and filter service tiers against the
  selected model and feature configuration for review and compaction steps.

## Testing

Added coverage for settings updates during paused turns, model changes,
previous-model compaction, review-model defaults, and service-tier filtering.

GitOrigin-RevId: fca4b3b305379f1bf26a5cac75ee0dd1747ac1f6
2026-08-25 17:41:22 +00:00
jif 7c6eb0eef1 Scope stop hooks for memory consolidation (#40587)
## Why

Background memory consolidation must honor managed completion policies without
running project completion checks or sending user-facing notifications.

## What changed

- Add a memory-consolidation stop target that selects managed policy hooks and
  executor-scoped cleanup while excluding user, project, session, and plugin
  stop hooks.
- End memory consolidation with an error when a managed stop hook rejects it,
  and propagate that turn error into agent status.
- Disable legacy completion notifications for the background memory worker.

## Testing

- Cover hook selection, executor cleanup, managed rejection, agent error
  status, and notification suppression.

GitOrigin-RevId: be350679ddcde2dde8a408b0972f08facff01a37
2026-08-25 10:29:26 +00:00
Won Park 8476434a97 Harden internal Guardian session isolation (#40497)
## Why

Internal Guardian reviews must remain independent of parent-session customization while still honoring managed execution and environment restrictions.

## What changed

- Apply the restricted Guardian session path to both internal and subagent review sources.
- Start internal reviews without inherited user instructions, extensions, MCP servers, or multi-agent behavior.
- Preserve managed execution policy and selected environment constraints while exposing only the allowed review tools. Honor shell feature restrictions across every selected environment and retain code mode when enabled.

## Testing

- Cover internal review isolation from parent state and managed instructions.
- Cover optional tool filtering, managed shell restrictions, code mode, and secondary environment profiles.

GitOrigin-RevId: edbf7d8a6fe3ee8213a1c75bc8ec331a1cf49006
2026-08-24 23:23:18 +00:00
Eddie Chen 49880081fe Add root turn IDs to turn and tool analytics (#40486)
## Why

Analytics need to associate subagent activity with the top-level turn that
caused it, while avoiding a stale association when steering makes that origin
ambiguous.

## What changed

- Add `root_turn_id` to turn and tool-item analytics events.
- Read the root from live turn metadata when each event is built, reporting
  `null` when the root is unknown or ambiguous.
- Propagate the metadata through control-tool and code-mode completion facts so
  correlated and background tool events retain the correct root.

## Testing

- Cover root IDs for top-level turns, subagent turns, correlated code-mode
  calls, and tool events emitted after turn completion.
- Verify that queued and subsequent events report `null` after the live root is
  invalidated.

GitOrigin-RevId: 42fd76541c3c716cec74787d47b50adb2d22e6d2
2026-08-24 22:30:59 +00:00
Krish Chainani 8b61c50ebe Run allowlisted executor plugin stop hooks (#40009)
## What changed

- Discover inline hooks from executor-provided plugin manifests and accept only
  the bundled Computer Use `Stop` hook for `node_repl.turn_ended`.
- Add the accepted hook to the current step's hook engine with
  executor-scoped provenance.
- Run executor-scoped hooks in the background after regular stop-hook
  decisions, without delaying turn completion, applying control effects, or
  exposing them in hook summaries. Skip them when regular hooks only block
  continuation.

## Testing

- Added coverage for manifest filtering and option preservation.
- Added coverage for background execution, interaction with regular stop
  hooks, and non-blocking turn completion.

GitOrigin-RevId: c0f75f83a23ba8a83bb3a5d5997c4882aacc32d6
2026-08-21 21:01:59 +00:00
Brandon Zhang 27c05a52e0 Include context window IDs in response metadata (#39847)
## What changed

- Add `context_window_id` to turn metadata for sampling and compaction requests, using the UUID for the model-visible context window.
- Keep the ID stable for requests in the same context window and advance it after compaction, independently of the existing thread-generation `window_id`.
- Reserve `context_window_id` so client metadata cannot override it.

## Testing

- Cover HTTP and WebSocket metadata, remote compaction, token-budget context alignment, resume behavior, and post-compaction window changes.

GitOrigin-RevId: 55cceea377ad509c8dd96ad668f2b5703ea46a63
2026-08-21 03:31:56 +00:00
rka-oai f20b63e85c Use Responses compaction for Amazon Bedrock (#39825)
## What changed

- Switch Amazon Bedrock remote compaction to `compaction_trigger` items sent through `/v1/responses`.
- Remove the legacy dedicated compaction protocol from provider capabilities.
- Default remote compaction to unsupported and explicitly enable it for OpenAI, Azure Responses, and Amazon Bedrock providers.

## Testing

- Cover manual and automatic Amazon Bedrock compaction requests and their compacted history.
- Verify remote-compaction capabilities for configured and Amazon Bedrock providers.

GitOrigin-RevId: cf9ba314595367518b37a85f334f0c105ce47e45
2026-08-21 00:53:52 +00:00
philn-oai e396ef3fc1 Add cwd-relative turn diff paths (#39625)
## What changed

- Add the opt-in `cwd_relative_turn_diffs` feature.
- When enabled, render turn diff paths relative to each selected environment's
  working directory instead of the detected Git root.
- Preserve repository-root-relative paths when the feature is disabled.

## Testing

- Cover enabled and disabled behavior for nested working directories across
  supported originators.

GitOrigin-RevId: 8b4908706a82b01f33ccd035b0945f9a29fce26f
2026-08-20 06:52:01 +00:00
rka-oai fb356f3d2c Add async delivery metadata to agent messages (#39312)
## What changed

- Add an optional `delivery` field to agent message events and app-server
  `agentMessage` items, with `"async"` identifying a user-visible message sent
  without ending the current turn.
- Preserve the marker through legacy event conversion, thread history
  materialization, replay, and generated JSON and TypeScript schemas.
- Require delivery of async agent message completion notifications on the
  in-process app-server transport.

## Testing

- Cover preservation of async delivery metadata in thread history.
- Cover delivery-required classification for async agent messages.

GitOrigin-RevId: f92ba25de1293ee271404badfdf0d117d6530329
2026-08-18 23:05:52 +00:00
guinness-oai 13dfaab446 Surface interactive requests in realtime conversations (#39113)
## What changed

- Mirror execution, permission, and patch approval requests into active realtime conversations with a prompt to review them in the app.
- Mirror user-input and elicitation requests with a prompt to respond in the app.
- Include the serialized request event so the realtime handoff retains the request details.

GitOrigin-RevId: 56fefa70b9ecaaad2fcf83592f20e8b4514a004a
2026-08-17 23:54:51 +00:00
joeflorencio-openai 00f6a8a60e Keep active-turn model settings stable across updates (#38785)
## Why

Thread settings can change while a turn is still active, including between sampling requests. Those updates should apply to the next turn instead of changing the model configuration partway through the current turn.

## What changed

- Snapshot the model, reasoning settings, service tier, approval settings, and model-attributed telemetry in `StepContext`.
- Build prompts, model requests, world state, tracing, and startup prewarms from that step-scoped snapshot.
- Share immutable model metadata between turn and step contexts.

## Testing

Added an integration test that pauses an active turn, updates its thread settings, and verifies that all requests in the active turn retain the original settings while the next turn uses the updates.

GitOrigin-RevId: c51352218e463dcc9b7338566d16742292c49fbf
2026-08-15 16:58:08 +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
sayan-oai 535795f7d1 Centralize turn environment selection state (#38461)
## What changed

- Store `TurnEnvironmentSelection` directly on each resolved `TurnEnvironment`
  instead of copying its environment ID, working directory, and workspace roots
  into separate fields.
- Update environment resolution, tool execution, approvals, and related tests to
  access selection state through the retained value.

GitOrigin-RevId: ab723d21e112b8af5bcaf1f772a5e376553312bc
2026-08-14 00:22:54 +00:00
sayan-oai 781445f7c6 Centralize thread environment selection state (#38423)
## What changed

- Make `ThreadEnvironments` the source of truth for live environment selections and use it when building configuration snapshots, permission profiles, per-turn configuration, and MCP refresh inputs.
- Keep settings previews side-effect free while applying accepted environment updates consistently to subsequent turns.
- Move `EnvironmentConfig` into `codex-protocol` so environment ownership no longer depends on a core-only type.

## Testing

- Cover environment previews and snapshots, accepted and rejected steer updates, and active-turn environment stability.

GitOrigin-RevId: 7bcf8fb5df1215f720e4e9c419541a51a871aa5e
2026-08-13 19:25:05 +00:00
Owen Lin da2803c73c Simplify queued user message admission (#38092)
## What changed

- Resolve user message admission when Core accepts the input as a new turn or a steer, without waiting for rollout persistence.
- Remove persistence- and hook-specific admission errors and task bookkeeping.
- Delete queued messages after Core accepts them for turn processing, including messages subsequently stopped by a prompt hook.

## Testing

- Update concurrent admission and queue service tests for acceptance-based admission.

GitOrigin-RevId: 7dcc311cfecd3d710733b420f2add546edb86153
2026-08-12 00:41:53 +00:00