## What changed
Remove `initial_messages` from `SessionConfiguredEvent` and the corresponding schema. Stop attaching replay history to startup events for resumed or rejoined threads, and remove `InitialHistory::get_event_msgs`.
## Testing
Update the MCP resume test to verify persisted tool events through `load_history`, and retain coverage that resumed requests include prior conversation items.
GitOrigin-RevId: 54b72bc44423c9b1ef2051891ef392c0ede32fe5
## Why
Unread queue-only messages prevented idle agents from unloading, and sending a
message to an evicted agent reloaded its session. Pending mail should not require
an idle recipient to occupy a loaded thread slot.
## What changed
- Retain queue-only mail in an in-memory runtime mailbox when a session unloads.
Queue new messages for locally evicted agents without reloading them or
starting a turn.
- Add `AgentControl::take_mailbox` and `AgentControl::watch_mailbox` so sessions
can consume retained mail and react to host mailbox notifications.
- Preserve message order across eviction, reload, and subsequent follow-ups.
Keep agents with outstanding durable sleep or turn-triggering mail loaded,
and discard retained mail when an agent is removed or its tree shuts down.
## Testing
Add integration coverage for retaining mail across eviction, sending while
unloaded, and ordering messages around a follow-up after reload. Add a host
controller test verifying that mailbox notifications wake `wait_agent` and
that the message is delivered exactly once.
GitOrigin-RevId: 4b55647f9ee9d1aaf3753f0688863bda81617327
## Why
Subagents spawned without forked history do not receive their parent's client-defined dynamic tools, preventing them from using those tools for delegated work.
## What changed
Add the disabled-by-default `multi_agent_v2_dynamic_tools` feature flag. When enabled, fresh V2 subagents inherit the parent's dynamic tool definitions and persist them in session metadata.
## Testing
Add coverage for V1 and V2 feature gating, tool persistence, and a fresh V2 subagent calling an inherited tool and receiving its result without inheriting parent history. Update TUI assertions and a snapshot to reflect blank-line separation and question text in recovered answers.
GitOrigin-RevId: a17a16bf5570f54ce922c1521cffb883022354f2
## What changed
Update `post_sampling_token_estimate_is_disabled_by_always_on_sinks` to register explicit event metadata with the subscriber and assert that its interest is `never`. This replaces the scoped default subscriber and global interest-cache rebuild with a direct check of the configured logging layers.
GitOrigin-RevId: 7c4ecd9c4adef148fd222394ad1ad326e8033d95
## 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
## Why
`codex exec` and the TypeScript SDK need a way to request a Cyber access program for an individual turn. Resuming with an API key can also leave compaction using a saved program that the current credentials cannot access.
## What changed
- Add `--cyber-access-program` with `standard`, `daybreak_blue`, and `daybreak_red`, and expose the equivalent `cyberAccessProgram` turn option and `CyberAccessProgram` type in the TypeScript SDK.
- Apply explicit selections to new, resumed, and prompted fork turns. Override automatic Daybreak selection for that turn without changing the saved choice. Reject the flag for review and forks without a prompt.
- Filter Cyber programs by the built-in OpenAI provider when starting turns and restoring previous-turn settings for compaction.
- Allow previous-model compaction fallback with API-key authentication, comparing both model and program so a program change can trigger a retry with the current pair even when the model is unchanged.
## Testing
Add CLI help, argument parsing, exec, and SDK coverage for per-turn selection and unsupported commands. Extend core tests for API-key compaction retries, denied explicit selections, custom-provider filtering, remote compaction, and WebSocket response reuse.
GitOrigin-RevId: 586f25b621924f3bc509c9a39b39fe2020ff9cd2
## 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
## Why
Background code-mode cells can continue running after a new turn starts. Looking up grants and strict review requirements through the active turn can make those cells lose their approved access or use another turn's permissions.
## What changed
- Keep turn-scoped grants and strict review state in `TurnContext`, retained by captured steps and background cells.
- Record permission responses against the requesting turn, including requests made after a later turn starts.
- Share live session-scoped grants across turns while keeping turn-scoped grants isolated.
- Use the originating turn's permissions and review requirements for tool execution, patches, MCP calls, file uploads, and terminal input.
## Testing
Add regression coverage for background cells retaining grants across turn changes, turn versus session scope, permission requests after the next turn starts, and strict review enabled while `write_stdin` is queued.
GitOrigin-RevId: 641502fbfd43553c2c42424d4aaa53ae8020c850
## 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
## What changed
Add `turn_extension_init` to thread startup and settings, and expose `WithTurnExtensionData` for turn input, recovery, and settings requests. Accept or reject data updates together with the request that supplies them.
Each new turn captures the configured data, including automatically started turns. Accepted steering updates the data for future turns while preserving the running turn's capture. Plain requests preserve the configured data; an empty initializer clears it. The data is not persisted and must be supplied again on cold resume.
Expose `current_turn_extension_data` to retrieve typed data only when the requested turn is still running and has not been cancelled.
## Testing
Add an integration test covering initial capture, accepted and rejected steering, inheritance by an automatic turn, clearing future data without changing the active capture, and rejection of lookups for an earlier turn.
GitOrigin-RevId: dd81eda3e32c7b62d783c772221858f4c0e68264
## What changed
Expose `ThreadManager::request_agent_tree_shutdown` and an `AgentTreeShutdown::wait` handle. Requests reject new starts in the shared runtime, cancel pending startup, and signal every session in that tree, including delegates outside the thread registry. Waiting covers admitted operations and session teardown, and reports cleanup or persistence writer failures. Other root trees remain unaffected; registry removal and persistence handoff remain caller responsibilities.
Retain tree membership through startup, cancelled spawn cleanup, residency eviction, and completion watchers. Release persistence resources before returning startup failures, and stop completion watchers during tree shutdown without reporting shutdown as a child result.
## Testing
Add coverage for tree isolation, children removed from the registry, delegate teardown, cancellation during required MCP initialization, aborted teardown, persistence writer failures, and successful retry after a failed resume releases its writer.
GitOrigin-RevId: a109e25adf23d20baf7595891099947050c6b6b5
## What changed
Capture each environment's effective Windows sandbox backend in the MCP runtime and include it as `useMxc` in sandbox state metadata for supporting servers. Use the server's captured environment selection, including local fallback, so the local MXC preference does not leak into attached environments. Default missing backend information to `false`.
## Testing
Extend tests to cover MXC metadata handoff, attachment isolation, local fallback without a selected local environment, and preservation of captured backend settings across runtime refreshes.
GitOrigin-RevId: f559bffeff383838334e5a475b1eab608d571fa1
## 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
## What changed
- Start a turn-owned notification receiver using the same remote board client and session identity as the board tools. Deliver post previews as agent messages and cancel the receiver when the turn stops, aborts, or errors.
- Wait for the initial connection attempt before inference, then retry failed or disconnected streams without waking idle agents.
- Skip malformed notices, mismatched recipients or turns, and oversized previews. Reject delivery to stale or canceled turns and bound remote notices to a 1 KiB context budget.
## Testing
Extend remote board tests to cover interrupted setup, connection failures, reconnection, invalid and stale notices, preview delivery into model context, and receiver shutdown after turn completion.
GitOrigin-RevId: c7c8d977fe74b7a5cb3bd8216dc0146f3f297e44
## Why
Resuming a thread rereads persisted history even when a preloaded snapshot is unchanged. Reusing that snapshot must still account for writes committed before writer ownership is acquired.
## What changed
- Carry optional, non-serialized revisions from stored history and model context through thread and agent resume paths.
- Validate the revision under writer ownership and reuse the shared snapshot when it matches. Reload stored history when the revision is missing or changed, while preserving explicit history overrides.
- Resolve the selected rollout before loading context, validate compressed snapshots before materialization, and enforce archived-thread restrictions on canonical paths.
## Testing
Extend coverage for snapshot reuse and late-write refresh in legacy and paginated histories, compressed rollout reuse, archived rollout symlinks, and cold child restoration that includes edits made after preloading history.
GitOrigin-RevId: e1fd7aee6ef2b30f41951e6b7e6b01c3e1420cab
## Why
A stored history snapshot can become stale before a resumed thread acquires writer ownership. Resume must include writes committed before that acquisition.
## What changed
Make `ThreadStore::resume_thread` return the replay context and use it for session restoration and metadata initialization. The local store refreshes stored snapshots while holding writer ownership, reads through the installed writer to honor explicit rollout paths, and releases the writer if loading context fails. Explicit histories without a canonical session header retain their override semantics.
## Testing
Extend resume tests to cover edits made after the initial snapshot in both legacy and paginated history modes, stale SQLite rollout paths, and preservation of explicitly supplied history.
GitOrigin-RevId: 2361f524d324c859f5dfcf2863320e2045736c2e
## Why
User goal edits and clears need to reach the model as user instructions, including when a thread is unloaded or history is being compacted. Automatic goal lifecycle updates must not supply user authorization.
## What changed
- Add optional `origin` values of `user` and `automatic` to `thread/goal/set` and `thread/goal/clear`, and mark TUI goal requests as user actions. Omitted origins do not supply user authorization.
- Persist explicit user objective, status, and clear instructions before changing goal state. Recording failures leave goal state unchanged.
- Coordinate goal edits with thread resume, reload stored history under writer ownership, and preserve edits accepted after the compaction input snapshot.
- Update generated protocol schemas and TypeScript and Python bindings.
## Testing
Add coverage for user edits and clears on loaded and unloaded threads, concurrent path-based resume, exclusion of automatic goals from user instructions, preservation of edits during compaction, and recording failures that leave goal state unchanged.
GitOrigin-RevId: d779d7c678a0c32a6ef76975dbe5b1400d1df245
## Why
Guardian does not consume skills, but host skill discovery can block its startup or turn creation when the primary executor is offline.
## What changed
Skip plugin and skill warmup during Guardian session initialization, and use an empty host skills snapshot when building Guardian turn contexts.
## Testing
Add a regression test that discovers skills for the parent session, disconnects its primary executor, and verifies that Guardian still reviews and denies a network permission request.
GitOrigin-RevId: 3c12129721c82cb78a3fa618d6b05faea631966a
## Why
Refreshing available model descriptions changes the `spawn_agent` tool schema. Keep model choices current while preserving tool definitions and earlier conversation input.
## What changed
- Add the disabled-by-default `model_catalog_in_context` feature for multi-agent V1 and V2. When enabled and model overrides are available, move model listings from tool descriptions into developer `<model_catalog>` messages.
- Include picker-visible, compatible models with descriptions, reasoning efforts, and service tiers in a listing bounded to 1,000 bytes. Point `spawn_agent` guidance to the latest catalog.
- Append updated listings only when their rendered content changes, restore missing listings, and invalidate retained catalogs when they no longer apply.
## Testing
Extend integration coverage for V1 and V2 with the feature enabled and disabled. Verify that catalog refreshes preserve tool definitions and prior input when enabled, append the updated listing once, and retain the existing tool-description behavior when disabled. Add a request-history snapshot for the V1 opt-in path.
GitOrigin-RevId: 67b11b4ad6c35f00d0e29edf6343bb0b4804d797
## What changed
Forward explicit `cyberAccessProgram` selections for the built-in OpenAI provider when both `api_key_cyber_access_programs` and `api_key_model_discovery` are enabled. The new feature defaults to disabled and can be toggled through the app server's experimental feature enablement API.
Reject explicit selections in OpenAI API-key sessions when either feature is disabled, instead of silently omitting the program. Leave entitlement and model restrictions to the server, and preserve the requested program when a server rejection triggers transport retries.
## Testing
Add app-server coverage for both feature gates, forwarding all three program values, and server rejection without fallback. Update core coverage for API-key resumes with disabled features and custom-provider omission.
GitOrigin-RevId: aa639afb841aa651b28f6f3f61198299372baa2d
## Why
Persistence and aggregate metadata budgets can omit tool-call observations that are still relevant to the current request, including tool result metadata needed during compaction.
## What changed
- Retain live direct-call observations separately from bounded persisted outputs, keyed by response item ID so reused call IDs cannot mix results.
- Restore those observations for sampling and compaction, pruning them only when sampling confirms they have left the history window. Preserve observations across shortened compaction retries.
- Remove aggregate prompt metadata caps while retaining per-call argument limits and the outgoing message size limit. Invalidate cell completeness when argument normalization truncates a recorded call.
## Testing
Add regression coverage for exhausted persistence budgets, reused call IDs and reversed completion order, shortened compaction retries, and completeness after argument truncation. Extend integration coverage for large Code Mode metadata, provider metadata support, and fork/resume behavior.
GitOrigin-RevId: 004082b69bff9cc8b05f36d1052057f163c67f07
## What changed
- Instrument MCP OAuth loads, saves, deletes, fallback cleanup, and refresh persistence with credential-free storage metrics, distinguishing configured policy from operations on a pinned store.
- Retain the originating client across background tasks, blocking storage operations, and session recovery, including recovery after startup found no credentials.
- Preserve typed keyring and lock error categories for metrics and remove raw error details from storage warnings.
## Testing
Add tests for policy and pinned-store outcomes, best-effort cleanup failures, wrapped lock errors, and successful or failed refresh persistence without duplicate observations. Add recovery tests that verify client attribution after anonymous startup and across threads with different originators.
GitOrigin-RevId: 5ea4528bcdcf500c8ffdf1d9c7008ea52112c726
## Why
Successful file fallback can hide a secure-store failure. Storage diagnostics need to distinguish that fallback from explicit file storage while keeping credential values out of error output.
## What changed
- Record credential load, save, delete, and fallback-file cleanup outcomes, including secure-store failures and fallback reasons.
- Preserve the initiating client's attribution across app-server requests, login tasks, sessions, turns, and blocking token persistence.
- Preserve storage error context and typed causes while redacting credential JSON data errors and payload-bearing keyring errors.
- Include sanitized fallback completion warnings in the default `codex login` file logs.
## Testing
Add coverage for storage policy and metric attribution, credential redaction, login diagnostics, concurrent app-server clients, device-code login, and token refresh during a turn.
GitOrigin-RevId: 6d5651c1718d184016fa77befa02aa528750b2ae
## Why
A subagent stopped by repeated Guardian denials remained interrupted without notifying its parent. The parent needs guidance to report the stop and wait for explicit user confirmation before continuing the blocked work.
## What changed
Forward `TooManyDenials` interruptions from `MultiAgentV2` children to their parents with Guardian-specific guidance to inform the user and avoid resuming or retrying until the user explicitly confirms. Preserve the child's `Interrupted` status and keep ordinary interruptions silent.
## Testing
Add integration coverage for parent notifications under `CircuitBreakAction::Strict`, silence under `CircuitBreakAction::Default`, and preservation of the child's interrupted status. Add a unit test ensuring long denial messages stay within the notification token limit while retaining the guidance.
GitOrigin-RevId: c9bb90827393068d9c92edaf0443a52e6ff57f9f
## Why
Stale thread capability roots can reconnect executors that are not selected for the current turn and shadow selected roots with the same ID.
## What changed
Filter out roots whose environments are absent from the turn snapshot before deduplication and readiness resolution. Update app-server capability tests to explicitly select the executor environments used by each turn, including after resume.
## Testing
Add regression coverage for full-access and workspace-write permissions, verifying that stale roots never connect an unselected executor, selected roots with matching IDs remain available, and turns with no environments expose no roots.
GitOrigin-RevId: afe375047cc5e8aad06df83334149510460dcdcc
## Why
Reloads can reject a valid thread configuration when the host configuration references a missing model provider that thread overrides replace. Concurrent cloud refreshes can also leave older policy snapshots or staged cache writes eligible for publication.
## What changed
- Compose preserved session layers before materializing refreshed configuration, retaining source-aware validation errors.
- Carry cloud policy revision bindings through config loading and role overrides. Reject stale MCP refreshes and disable enterprise MCP authorization when the policy is suspended. Retire the old policy when replacing or clearing its loader.
- Stage cloud cache writes in temporary files and publish them atomically only while their policy revision remains current.
- Route file reloads through the shared runtime refresh path, resolve paths relative to each config file, and preserve session settings. Restore host defaults for `secret_auth_storage` and `mcp_oauth_refresh_coordination` when explicit overrides are removed.
## Testing
Add regressions for thread-provided model providers, managed plugin and feature requirement refreshes, stale and suspended policy snapshots, superseded and retired cache writes, and file reloads preserving runtime settings and restoring feature defaults.
GitOrigin-RevId: 3280a2445eeb7074a142f720690bb0e4f01d174b
## Why
Steering and recovery requests can have different traces from the turn they join. Sampling can also finish while tools remain in flight. Tracing needs to preserve that correlation and distinguish sampling time from time spent waiting for tools.
## What changed
- Add `codex.turn_input` spans with `conversation.id` and the accepted `turn.id` for started, steered, and recovered turns.
- Add sampling, tool-blocking, and automatic and manual compaction spans with `codex.turn.phase` and conversation and turn IDs. End the sampling span before draining outstanding tools.
- Emit a `codex.mailbox_preemption` event when pending mailbox input preempts sampling.
## Testing
Add integration coverage for cross-trace steering and recovery, sampling and tool-blocking boundaries through a blocked dynamic tool, and completion or interruption. Extend mailbox preemption tests to check event emission with preemption enabled or deferred.
GitOrigin-RevId: 59893c2d463c02cbd05f5c332e5afc2351cc8b62
## Why
Configuration reloads must honor current managed restrictions without granting new enterprise MCP authority to existing sessions or overwriting newer configuration with stale snapshots.
## What changed
- Allow enterprise managed authentication only for configured MCP servers. Remove plugin `ema_auth` overlays from the schema and disable servers that still use them.
- Refresh ordinary MCP settings while preserving session settings and initially admitted enterprise registrations. Disable enterprise MCP when its registration changes, policy revokes access, or configuration loading or resolution fails; require a new session to restore enterprise authority.
- Publish refreshes only when their captured configuration is still current, with bounded retries in app-server reloads. Preserve updated managed restrictions even when a refresh is rejected.
- Finish processing other thread refreshes before `config/mcpServer/reload` reports a rejection, and document that an error can accompany partially applied changes.
## Testing
Add regression coverage for stale refreshes, bounded retries, managed-policy revocation, persistent enterprise disablement, ordinary MCP transport changes, and rejection of plugin enterprise authentication. Verify that denied enterprise servers receive no requests.
GitOrigin-RevId: 34429cd339c1ec152a2ee8697cb29f0cff5d9e5f
## Why
Turn error hooks receive only a `CodexErrorInfo` category, which omits backend metadata such as usage-limit reset times and rate-limit snapshots.
## What changed
Add borrowed `CodexErrorDetails` to `TurnErrorInput` and pass the original details through turn and compaction error notifications, alongside the existing error category.
## Testing
Add coverage for usage-limit metadata during sampling, manual compaction, and post-turn compaction, including preservation of completed answers and existing error events. Extend lifecycle tests to verify that an `InvalidImageRequest` retains its details when surfaced as `BadRequest`.
GitOrigin-RevId: 6f2058992126da0f69508ef51998e7581557deea
## Why
Providers with `model_catalog_url` could advertise bundled models absent from their catalog or reuse stale models after a failed refresh. Model metadata could also match by prefix or namespaced suffix instead of the provider's exact model ID.
## What changed
- Suppress bundled model fallback and merging for explicit provider catalogs, including when discovery is disabled.
- Require unique, non-empty catalog slugs and match metadata by exact model ID.
- Serialize explicit catalog refreshes and invalidate in-memory and persisted catalog entries after fetch failures, preventing stale cache reuse.
- Report a clear configuration error when startup has no available model, while allowing an explicitly configured `model`.
## Testing
Add regression coverage for exact model matching, hidden catalog entries, cache invalidation across restart and recovery, invalid slugs, and empty-catalog startup with and without an explicit model. Update app-server model-list expectations and verify the TUI configuration error.
GitOrigin-RevId: ac323a94178e3c48df8d4a159a86fdbac20b03a2
## What changed
Move content-filter guidance recording from the sampling loop into
`handle_response_stream_error`. Pass `StepContext` so the handler can resolve
model-specific guidance and record it before retry decisions for sampling
requests. Keep remote compaction excluded from guidance recording and update
callers and the existing retry-deadline test to use the new signature.
GitOrigin-RevId: 030bde1d4f1597fc4f3971ad837478e5304c5aa9
## Why
Responses stopped by `content_filter` previously followed generic stream-error retry handling without guidance about the block. Retried requests need instructions to explain the limitation and offer a permitted alternative.
## What changed
- Recognize content-filter failures as a distinct error while preserving retryability and the existing public error representation.
- Append developer guidance to the conversation after each sampling block, retaining delivered output and using the existing retry handling.
- Support a model catalog override through `model_messages.content_filter_guidance`. Missing, null, blank, or values exceeding 512 UTF-8 bytes fall back to bundled guidance.
- The bundled guidance tells the model to avoid reproducing or working around blocked content, briefly explain the limitation, offer a permitted alternative, and continue unrelated authorized work.
## Testing
Add a scenario and request-history snapshot covering repeated blocks, catalog guidance on each retry, retained partial output, and eventual completion. Add error-mapping assertions for retryability, public error classification, and error text.
GitOrigin-RevId: ba0c201b3a5a3acf5cfa971332027240e0690225
## Why
Sessions can share an analytics client while using different product SKUs. Analytics requests need per-thread attribution to reflect each session's configuration.
## What changed
- Register each thread's `apps_mcp_product_sku` and send `X-OpenAI-Product-Sku` on analytics requests grouped by product, without changing authentication.
- Send product groups concurrently while preserving event order within each group.
- Flush buffered tool events before changing or evicting their product attribution, and clear attribution when a session disables analytics.
- Bound retained thread registrations and concurrent product groups. Drop attribution when registration cannot be queued or a batch exceeds the product-group limit.
## Testing
Add tests for shared-session attribution, clearing attribution on resume with analytics disabled, buffered events across product changes, queue overflow recovery, registry limits, and concurrent delivery with preserved event order.
GitOrigin-RevId: 25a98788cd44d4a85f97ee87aadf3a6909607f6a
## Why
Usage-limit failures during manual or post-turn compaction did not notify turn lifecycle extensions, leaving them unable to react to these errors.
## What changed
Emit the turn error lifecycle notification for `UsageLimitExceeded` in both compaction paths. Preserve the completed answer after post-turn compaction fails, and avoid duplicating the error already emitted by manual compaction.
GitOrigin-RevId: f31155e5991d762a0779030921aedbd1f818cfbb
## Why
An offline secondary executor can stall Guardian approval revalidation while diff display setup searches for Git roots, even when the review concerns another executor.
## What changed
Use each environment's working directory as the diff display root for Guardian sessions, regardless of `CwdRelativeTurnDiffs`, so this setup does not wait for remote Git discovery.
## Testing
Add a regression test that withdraws authorization during a pending approval and disconnects an unrelated secondary executor. Verify that Guardian retries with the updated authorization and denies the request.
GitOrigin-RevId: 3661759119935c68539e9eab8df2e862987d0d92
## Why
Spawning a subagent before an environment was ready dropped the starting environment from the child's selections. The child needs to retain that environment and receive its configuration or failure when preparation finishes.
## What changed
- Pass the spawning step's environment snapshot to children, preserving ready and starting environments and configuration ownership.
- Propagate the original owner's first configuration result through descendants, applying each session's validation and leaving independently configured children unchanged.
- Preserve pending configuration waits across executor retries. Later owner updates do not overwrite the child's inherited configuration.
## Testing
Add regression coverage for fresh and forked subagents, readiness through turn-settings updates, failure propagation to children and grandchildren, child configuration precedence, and executor retries. Add a request snapshot showing a child using `wait_for_environment` before execution tools become available.
GitOrigin-RevId: a2afed740391f88d195596cfe28e418332b60d43
## Why
Message-board notifications delivered as ordinary injected input could cause an agent to sample again after its final answer.
## What changed
Route notifications through the current turn's mailbox as `InterAgentCommunication`, checking atomically that the turn is running and still accepts mailbox delivery. Skip notifications when delivery is closed. Accepted notifications may remain in conversation history for a later turn without reopening the finalized answer or starting a new turn.
Report pending inter-agent communications as mailbox activity, while preserving priority for steering input.
## Testing
Add regression scenarios for notifications arriving before and after finalization, asserting no extra sampling and the expected notification presence in the next turn. Extend input-queue coverage to recognize already-pending mailbox activity.
GitOrigin-RevId: 41d786041b4eccc5ccf926f2f190473676a3ac2f
## What changed
- Add `CodexThread::prewarm_with_history()` to prepare a WebSocket response with existing conversation history and executed-tool metadata using `generate: false`. The next turn can reuse the prepared response when its prompt extends that history and request settings match.
- Keep startup prewarming and `prewarm()` on the existing empty-input path, and skip preparation when the cached WebSocket is ready.
- Label prewarm telemetry by input mode and WebSocket continuation metrics with `after_prewarm`.
- Rename `persistent_mode_enabled` to `persistent_execution_enabled` without changing its behavior.
## Testing
Extend WebSocket coverage to check history preservation during reconnect prewarming, reuse of the prepared response on the next turn, and omission of `previous_response_id` for a non-prefix prompt.
GitOrigin-RevId: c7731541b11b2e6668027a47f37f696213268c66
## Why
Guardian denial-limit interruptions lack a structured error identifying the cause. Make this opt-in because older clients may not recognize the new error in shared history.
## What changed
- Add `auto_review.circuit_break_action = "strict"` to attach `TooManyDenials` to the interrupted turn. The default leaves the error unset.
- Expose `turn.error.codexErrorInfo = "tooManyDenials"` in app-server notifications and saved history, and update protocol schemas and generated types.
- Preserve the warning, denial limit, and interrupted status without emitting a separate error notification.
## Testing
Add an integration test covering omitted, `default`, and `strict` settings, checking warnings, live turn errors, the absence of separate error notifications, and persisted turn errors after restart.
GitOrigin-RevId: cb5d379e64728d4d7b9daf834c13756e4a70db31
## Why
Inspecting one MCP server should not require full-inventory discovery or a separate connection when a thread already has one.
## What changed
- Add optional `serverName` to `mcpServerStatus/list` and update generated schemas and bindings.
- With `threadId`, apply any pending runtime refresh and reuse the thread's current connection and tool catalog, waiting only for the selected server during discovery.
- Without `threadId`, create a discovery connection for only the selected server. Unknown names return an empty page; omitting `serverName` preserves full-inventory discovery.
## Testing
Add a regression test covering both status detail modes, checking that tool metadata is preserved and that status reads do not repeat `initialize` or `tools/list` requests.
GitOrigin-RevId: 1b361b67a993ab45cae493e88c146f50cfaf753a
## Why
Guardian reviews need to retain original evidence across parent compaction when checkpoint reuse is disabled, including after resume and rollback.
## What changed
- Make disabling `guardian_reuse_parent_compaction` select a bounded, independent review transcript while preserving thread-owned authorization.
- Keep synchronous reviews and asynchronous scoring independent of parent checkpoints, and allow reviewer sessions to continue using transcript deltas across parent compaction.
- Persist transcript entries with rollback provenance while remaining compatible with older checkpoint readers and metadata-free checkpoints.
- Exclude compaction output from the transcript and synthetic summaries from rollback turn boundaries. Use acceptance ordering to remove rolled-back evidence even when persistence order differs.
## Testing
Add regression coverage for reviewer continuity, evidence retention across compaction and resume, rollback after local and remote compaction, and checkpoint serialization compatibility.
GitOrigin-RevId: fea0852cc5537372854f4a34bacb6bfda6d13e88
## Why
Guardian needs the assistant's delivered question to interpret a user's reply. Messages sent through nested Code Mode tools must remain available as review context, and approvals based on earlier context must become stale when a new delivery is confirmed. Assistant messages do not themselves grant authorization.
## What changed
- Capture successful hosted User Messaging sends after input rewriting and before result callbacks or post-tool hooks.
- Retain bounded delivery text in acceptance order, persist it through cancellation and shutdown, and preserve it across compaction and resume with an older-client-compatible rollout encoding.
- Track assistant review context separately from user authorization, invalidating pending and cached approvals for both local and worker reviews when that context changes.
- Associate retained deliveries with the correct instruction or communication boundary during rollback.
## Testing
Add coverage for nested messaging review context, fast replies and communication ordering, shutdown persistence, stale local and worker approvals, rollout compatibility, and rollback retention.
GitOrigin-RevId: 1bc1ec9e8947560b99f4b39a42ddb9f2099f519b
## What changed
Add `Features::persistent_mode_enabled` and use it for persistent instructions and current-time reminder defaults. Enablement still requires `ReasoningEffort::Persistent`.
Change `PersistentModeState::new` to accept an explicit enablement boolean, separating instruction rendering from reasoning-effort selection.
## Testing
Update persistent-context tests to use boolean enablement while preserving coverage for instruction replacement, removal, deduplication, and retained history without a snapshot.
GitOrigin-RevId: f412b90d783438d1526956a56c11b9e66385ee0e
## Why
Steering an active WebSocket response previously dropped the connection and
resent the full history. Draining the response preserves the connection and
allows the follow-up request to continue with `previous_response_id`.
## What changed
- Drain WebSocket responses when steering. For models using
`use_responses_lite`, first send `response.interrupt` with
`mode: "discard_partial_items"` once the response ID is available.
- Treat `response.incomplete` with reason `interrupted` as completion with
`end_turn: false`, preserving token usage and allowing the turn to continue.
Other incomplete reasons remain errors.
## Testing
Update the steering integration test to cover completed and discarded reasoning
items, asserting connection reuse, incremental follow-up input, and token usage
from the interrupted response.
GitOrigin-RevId: bc714b713e797cf1a67e3b26ffbf7664cb92eb33
## 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
## Why
Compaction using the previous model can inherit the current turn's access program, producing a model/program pair that the server rejects.
## What changed
- Persist `cyber_access_program` in previous-turn settings and restore it during rollout reconstruction.
- Use the previous turn's access program when compacting with its model, preserving an absent program instead of inheriting the current selection.
- Pass the selected access program into local compaction prompts.
## Testing
Add regression coverage for local and remote compaction after model switches, including resume, fork, rollback, and compaction checkpoints. Verify fallback and subsequent sampling use the current model/program pair, and API-key sessions do not inherit access-program authorization.
GitOrigin-RevId: eaa7162e8711446dc9eb1bfff194e711abd54d7e
## Why
Cached executor MCP declarations must not gain access to host environment credentials when reconnecting to an executor that lacks environment credential resolution.
## What changed
- Retain credential policy through MCP registration, catalog materialization, and runtime setup. Derive selected HTTP plugin policy from the plugin's source environment.
- Reject host bearer-token fallback and transports requiring host environment headers or helpers for executor-owned declarations. Preserve host-configured credential resolution.
- Include credential policy in connection and tool catalog cache identities, and avoid reading host environment values for executor-only identities.
## Testing
Add regression coverage for legacy executors, cached declarations after a capability downgrade, policy preservation through catalog rebuilds, and connection and tool cache isolation.
GitOrigin-RevId: 2d5b85e4e98668c637c538cbe528914c936bd831
## Why
With `instant_interrupt` enabled, new user input should reach the model without waiting for an unfinished response or stream retry backoff.
## What changed
- Interrupt request setup, response streaming, and retry backoff when user input arrives, then continue the turn with the queued input.
- Reset the WebSocket connection and continuation state after preemption so the next request sends full history.
- Drain pending input after a preempted step even when compaction leaves the estimated context above the token limit.
## Testing
Add regression coverage for WebSocket reconnection with full history, interrupted retry backoff, input queued during compaction, and yielding running code-mode calls. Add a streaming scenario that preserves completed commentary while excluding unfinished assistant text from the replacement request.
GitOrigin-RevId: b7e42b6afffbf87da15af05e0850af8579d1353f
## Why
User input queued during a long-running code-mode `exec` or `wait` can remain pending until the call returns. Allow these calls to yield early so Codex can receive the new input while the cell continues running.
## What changed
- Add the `instant_interrupt` feature flag, disabled by default.
- Watch queued user input for each sampling request and pass a shared preemption signal to code-mode `exec` and `wait` calls, including calls received later in the same response.
- Yield running cell IDs without stopping the cells, allowing subsequent `wait` calls to collect their results.
## Testing
Add integration coverage for enabled and disabled behavior, repeated steering during `exec` and `wait`, input deferred during compaction, and later calls in the same response. Verify that direct tool results and queued user messages are preserved, and add a scenario snapshot showing continuation after a cell yields.
GitOrigin-RevId: e33466a980d98e113bfb6c6a45c59de0cc7b3c1d
Owen Lin