## Why
Queued prompts and `request_user_input` answers can be persisted in a different
order from when the host accepted them. Using persistence order can therefore
reorder retained instructions during replay or apply the wrong rollback
boundary.
## What changed
- Reserve an acceptance sequence for user prompts and elicitation responses
when retained thread context is enabled.
- Persist that sequence with user-message metadata and verified answers, and
use it when ordering, bounding, replaying, and rolling back retained context.
- Keep events without acceptance metadata compatible with legacy recording
order.
## Testing
- Cover delayed queued-input recording, rejected prompts, checkpoint replay,
compaction, rollback migration, and legacy retained-context behavior.
GitOrigin-RevId: 42d24465cc2033bc51c0c2da79d3c0884a8ad98a
## Why
Guardian review needs the original user instructions alongside verified answers, even after compaction, resume, or rollback. A summarized or legacy checkpoint cannot establish a complete instruction history.
## What changed
- Capture text user messages in retained context when `guardian_thread_context` is enabled, preserving their message identity through compaction and replay.
- Keep retained instructions and verified answers in arrival order, with independent size and completeness tracking for each family.
- Roll back retained evidence at the exact user-message boundary, including steers that share a turn ID, and apply the same semantics during rollout migration.
- Treat checkpoints without retained user messages as incomplete while preserving their verified answers.
## Testing
- Cover capture, compaction, resume, rollback, child forks, legacy checkpoints, storage bounds, and rollout migration.
GitOrigin-RevId: 72706731b54e5ac8645db61b446d2401afb42235
## 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
## Why
A remote process can finish while a network policy review is still pending. Normal process cleanup should withdraw that review without turning the completed command into a review failure or losing its output.
## What changed
- Record whether a network policy request was withdrawn because the process finished, was cancelled, lost its executor connection, or timed out.
- Treat normal process completion as cleanup while retaining fail-closed behavior for other cancellation causes.
- Preserve explicit network denials before policy persistence so cleanup cannot replace the reported call outcome.
## Testing
Add an integration test that completes a remote process during a pending network review and verifies that the command reports its successful exit and output without approving the withdrawn request.
GitOrigin-RevId: 7f42d75631ee29eba43bf04cc953eea490f553fc
## 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
## What changed
- Add the stable, default-on `unified_exec_tty` feature.
- When disabled, omit `tty` from the `exec_command` tool schema and reject calls that explicitly request a TTY.
- Continue to support commands that omit `tty` or set it to `false`.
## Testing
- Add unified exec coverage for the disabled tool schema, explicit TTY rejection, and non-TTY command execution.
GitOrigin-RevId: c5934fe9db1423170b891c7ecd0188792cd03f42
## What changed
- Add the macOS-only `allow_symlinked_codex_home` user setting. When enabled in the execution host's `$CODEX_HOME/config.toml`, writable roots at or beneath that home may traverse symlinks.
- Keep the exception disabled by default and prevent project config, command-line overrides, and ignored user config from enabling it. Other writable roots remain subject to symlink checks.
- Propagate the resolved setting through local, interactive, and exec-server sandbox paths, including `CODEX_HOME` aliases.
- Explain the opt-out in symlink rejection errors, including that it trusts targets outside `CODEX_HOME` and targets that change between commands.
## Testing
- Add macOS coverage for shell commands, patches, filesystem helpers, process execution, and interactive startup with enabled, disabled, aliased, and out-of-scope homes.
GitOrigin-RevId: 99fcdf611200c9e1b7713cf06f7fdea5bfa7f089
## What changed
Remove guidance that told `send_message_to_user_async` to request missing
information, preferences, constraints, clarification, or approval. Keep the
tool focused on critical blockers, direction-changing findings, and replies to
user questions or status requests during ongoing work.
GitOrigin-RevId: c764ffc2b9c14e37467bd25ee3ed07407d550290
## What changed
- Add the `codex-attachment-store` crate with storage-neutral attachment metadata, references, errors, and an asynchronous persistence interface.
- Provide an inline implementation that preserves attachment bytes as media-typed base64 data URLs.
- Inject the store into `ThreadManager`, expose it to consumers, and retain inline storage as the default for existing entry points.
## Testing
- Verify inline storage round-trips binary, text, PNG, and JPEG data.
- Verify attachment debug output redacts URLs while retaining file IDs.
GitOrigin-RevId: 7688dcd3c89d7540ed2398f3e7c63881fbfcda97
## What changed
- Add `PromptImageMode::HIGH_DETAIL` and `PromptImageMode::ORIGINAL_DETAIL`
constants with the standard resize limits.
- Use the shared modes during core image preparation instead of defining the
limits locally.
- Cover the dimension and patch budgets for both detail modes in the image
utility tests.
GitOrigin-RevId: 27fdc77719f23d2e8f1060886576b3be843a8491
## Why
Guardian can reuse an opaque parent compaction only when the checkpoint's
recorded producer hash matches the scoring model. Missing or incompatible
provenance must not allow an asynchronous score or cached approval to omit
compacted context.
## What changed
- Record the producing model's compatibility hash on compaction history items
and expose the latest checkpoint hash through conversation snapshots.
- In thread-owned context mode, fail closed and require synchronous review when
checkpoint reuse is enabled but the latest checkpoint is incompatible or has
unknown provenance.
- Reject incompatible checkpoints in the sampler while preserving the legacy
context behavior.
## Testing
Add coverage for local, remote, resumed, compacted, and rolled-back histories,
including matching, mismatched, missing, and empty compatibility hashes.
GitOrigin-RevId: 1ce959cb276edc602fd7dfa442133c7fe5e53857
## Why
Guardian authorization reviews need host-verified `request_user_input` answers to
remain aligned with thread history across compaction, resume, and rollback.
## What changed
- Record verified question-and-answer pairs in retained thread context when
`GuardianThreadContext` is enabled, and source both local and delegated Guardian
review evidence from that context.
- Preserve complete answer records within the evidence budget. Mark authorization
context incomplete when records are unavailable or oversized, and defer cached
fast approvals in that case.
- Keep the existing runtime-only answer path for threads that do not enable the
thread-owned context.
## Testing
- Cover answer retention through compaction and resume, removal after rollback,
source-call eviction, oversized answers, delegated review evidence, and legacy
behavior.
GitOrigin-RevId: 01f6f4541515434a0d00fdde318ff59f2f8bd2eb
## What changed
- Pass the selected executor's `PathUri` through user command approval requests instead of converting the working directory to a host-native path.
- Make native-path permission helpers explicitly local and require executor-aware permission transforms to receive a `FileSystemSandboxPolicyContext`.
- Cover approved remote commands to verify that the approval request reports the selected remote working directory and execution resumes after approval.
GitOrigin-RevId: 0aa3042f2d7594af6c7d0abf31d7f99cf67d5556
## What changed
- Add the creation-time `originator` to app-server thread responses and `thread/started` notifications.
- Persist originators in thread metadata and preserve the first recorded value across list, read, resume, rollout, and SQLite paths.
- Add an `originators` allowlist to `thread/list` for hosted backends. The local app-server accepts omitted or empty values but rejects nonempty filters.
## Testing
- Cover originator persistence across server restarts and list, read, and resume requests.
- Verify local `thread/list` handling for nonempty, omitted, null, and empty originator filters.
- Heap-allocate large TUI dispatcher futures in affected tests to keep them within the Windows test-thread stack.
GitOrigin-RevId: 73ad6e61b8fdd2d93a22d1f971723a19ec7be3fc
## What changed
- Add `originator` to plugin measurement inputs and emitted analytics events.
- Populate it from the turn originator when recording plugin measurements.
- Extend analytics tests to verify the originator is preserved in event payloads.
GitOrigin-RevId: 9998f829760222fc690dc8a10324794b5da27220
## What changed
- Add `application.network` managed requirements with exact-domain allow/deny
rules, default-enabled behavior, domain normalization, and normal managed TOML
precedence.
- Return the policy separately from agent network requirements through the
experimental `configRequirements/read` API. The endpoint reports the policy
but does not enforce it.
- Export the new protocol types and schemas, and document the TOML format.
## Testing
- Cover parsing, validation, normalization, layered precedence, and cloud
requirement composition.
- Verify app-server responses, defaults, separation from agent policy, and
rejection of invalid domains.
GitOrigin-RevId: 563eeed296d9748430b819797204a44a8db92f3b
## 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
## What changed
- Set `CODEX_VERSION` to the running package version in user shell and unified
exec environments, overriding stale shell-policy or snapshot values.
- Add the reserved `codex_version` field to MCP turn metadata.
## Testing
- Cover version propagation through shell policies, shell snapshots, unified
exec, and MCP metadata.
GitOrigin-RevId: 7846c70d668176530b1f467293b749dc9ddd79a0
## Why
Patch targets can use a different path convention from the Codex host, so
host-native path conversion can misclassify writable roots and requested
permissions.
## What changed
- Evaluate patch targets as `PathUri` values with the active filesystem policy
context, including workspace roots and the executor's path convention.
- Distinguish executor-managed sandboxing from local platform sandboxing when
deciding whether a patch can be auto-approved and how to normalize additional
write permissions.
- Make full-disk and special-path policy checks honor the selected executor's
Windows or POSIX convention.
## Testing
Add coverage for Windows executor URIs, full-disk policy aliases, remote patch
permission requests, sandbox availability, and owner-provided workspace roots.
GitOrigin-RevId: 1a054ea443efd342623c67432762f85c53d20c15
## What changed
- Add the under-development `features.context_management.experimental_mode` configuration and schema.
- For eligible ChatGPT Plus, Pro, and Pro Lite sessions using the Codex backend, enable token-budget context, history notes, and the `new_context` tool when experimental context management is configured.
- Keep the feature disabled for custom providers, provider credentials, non-Codex endpoints, and temporary structured threads.
## Testing
- Cover subscription eligibility, backend restrictions, configuration resolution, token-budget guidance, and temporary-thread overrides.
GitOrigin-RevId: 19402af19c4d69e09a7cdb3687ee7d233f6a562a
## Why
Directly deserializing the flattened `RolloutLine` envelope can reject nested
decimal values, preventing affected paginated sessions from resuming.
## What changed
- Add canonical string, byte, and reverse-scanner helpers that decode rollout
records through `serde_json::Value` before decoding the flattened item.
- Route rollout readers across session discovery, history, migration, search,
thread storage, and transcript previews through those helpers.
- Remove `Deserialize` from `RolloutLine` so new readers cannot bypass the
canonical persistence decoder.
## Testing
Add coverage that resumes a paginated rollout after a token-count record with
a decimal rate-limit value and verifies that ordinal sequencing continues.
GitOrigin-RevId: 49abac1e0751c073daa5a93a840d8a483fd2d013
## Why
Exec-server process and network-policy events need stable attribution to the
launching tool call and executor without recording process payloads.
## What changed
- Add optional `ExecMetadata` to `ExecParams` and propagate the thread and tool
call IDs from unified exec.
- Emit bounded OpenTelemetry events for process start, spawn failure, sandbox
denial, and exit. Correlate them with the launch trace and registry-issued
executor identity while excluding arguments, paths, environment values,
output, and error text.
- Preserve launch attribution across long-running processes and reconnects, and
attach the same metadata to network-policy audit events.
- Keep the protocol backward compatible when metadata is omitted, and prevent
invalid trace headers from inheriting an unrelated active span.
## Testing
Add coverage for metadata serialization and propagation, lifecycle event
fields, trace relationships, reconnect behavior, spawn failures, sandbox
denials, and network-policy attribution.
GitOrigin-RevId: 7aa480a7289c73cb95e2c124c35500bb6f0d5084
## Why
When ordinary included usage is exhausted, eligible users need a way to keep working with the backend-provided Luna Reserve quota and return to their previous model once ordinary usage recovers.
## What changed
- Automatically switch eligible TUI tasks to Luna Reserve after an authoritative account usage read, preserving the prior model and reasoning effort per task.
- Restrict model selection while on Reserve to its supported reasoning levels, keep queued turns on the accepted model, and restore the saved model after a fresh read confirms recovery.
- Surface Reserve usage in the composer and `/status`, and expose `normalModelSlug` in app-server rate-limit snapshots so clients can use the associated model's display metadata without changing the request model.
- Poll account limits more frequently near exhaustion and retain compatibility with app servers that only accept the legacy null request parameters.
## Testing
- Added coverage for Reserve entry, queued-turn handling, task reconstruction, recovery authorization, account changes, model selection, usage rendering, and banner dismissal.
GitOrigin-RevId: 46a435e9531051f8cec3f6a46c4b9f7cfa479747
## What changed
- Add the `send_message_to_user_async` control tool for concise questions and updates during ongoing work.
- Emit messages as asynchronous agent items without ending the turn, and reject empty message text.
- Expose the tool only to root agents when the model catalog explicitly opts in, independently of `request_user_input_async`.
## Testing
- Cover catalog gating for root agents and subagents.
- Verify asynchronous item emission, tool output, empty-message validation, and continued turn execution.
GitOrigin-RevId: 8c6a7988a502f3dbc9eff7000913e85f0c596550
## What changed
- Add the default-off `windows_sandbox_service` feature and expose it through
app-server experimental feature enablement.
- When enabled, attempt service provisioning during elevated Windows sandbox
setup, using the effective network proxy ports and listener roles. Continue
through the existing setup path when the service is unavailable or the
configuration is unsupported.
- Report unsupported Codex home drives as service unavailability so the client
can fall back to the elevated helper.
## Testing
- Add a Windows-only unit test covering HTTP and SOCKS listener derivation from
the effective proxy configuration.
GitOrigin-RevId: 40b119cf51f84e5d0e94d6a0a8d5b506d1123526
## 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
## What changed
- Add the top-level `thread_unload_delay_secs` configuration key for the
app-server, with a 60-second default instead of the previous 30-minute
fixed delay.
- Allow zero-delay unloading and reject values that cannot fit in a
monotonic-clock deadline.
- Reset the unload countdown when a thread gains a subscriber or becomes
active, and preserve active turns even when the configured delay is zero.
- Document that unloading ephemeral threads discards their in-memory state.
## Testing
- Cover configuration parsing, defaults, overrides, and overflow rejection.
- Cover unsubscribe, resubscribe, delayed and immediate unloading,
notifications, persisted-thread resume, and WebSocket disconnect behavior.
GitOrigin-RevId: edd46f6b49bbdafbf606bff74378e8f569e95977
## Why
Permission requests can originate in an environment whose path convention differs from the Codex host. Requiring the request working directory to be host-native prevents those requests from reaching interactive approval.
## What changed
- Carry the request `cwd` as a `LegacyAppPathString` through core events and the app-server protocol so the target environment's native representation is preserved.
- Convert the preserved value back to a `PathUri` when materializing permission grants.
- Regenerate the JSON and TypeScript protocol schemas for the updated `cwd` type.
## Testing
- Exercise the app-server permission request round trip with a target-native working directory across host path conventions.
- Verify remote permission requests retain their environment working directory and granted access still unblocks later execution.
GitOrigin-RevId: 97dc9707d9048883d20bab1b69c93871039f3188
## What changed
- Limit elevated interactive setup to provisioning sandbox identities and proxy settings; do not include caller workspace or ACL roots in its payload.
- Always run the non-elevated setup refresh during sandbox activation so the current permission profile supplies read, write, and deny paths, including when provisioning was completed previously.
- Add an `interactive-provision` setup mode that performs provisioning without the full ACL setup path.
## Testing
- Verify elevated provisioning payloads contain no caller ACL roots and use the new mode.
- Verify the setup helper accepts `interactive-provision` payloads.
GitOrigin-RevId: 081b1b43abd2388773c230fb35da82896e7be2ea
## Why
A steer creates a new instruction boundary without creating a new turn ID. Retained `request_user_input` answers that share that turn ID must not all be discarded when only the later steer is rolled back.
## What changed
- Associate retained answers with their originating function call and instruction boundary.
- Keep answers whose source calls survive rollback, including through compaction, resume, and rollout migration.
- Remove answers whose source calls belong to the rolled-back instruction.
## Testing
- Cover live rollback and resume with answers before and after a steer.
- Cover rollout migration when delayed answers share a turn ID but originate from different calls.
GitOrigin-RevId: 225e7eb7b6a3ae5a5e7fa6fcb80cb7d259db90d0
## Why
Compaction replaces the model history window, but host-verified user answers may
still be needed for the rest of the owning thread.
## What changed
- Add bounded, model-invisible retained context for verified
`request_user_input` answers.
- Persist retained context in rollout events and compaction checkpoints, and
restore it when a thread resumes.
- Remove answers when their source turns are rolled back, and prevent
parent-local answers from being inherited by child forks.
## Testing
Add lifecycle coverage for deduplication and size limits, compaction and resume,
rollback, rollout migration, and child forks.
GitOrigin-RevId: c32900d3e8120a9899b71f0576683acd67a4b898
## What changed
- Skip Guardian prewarming and asynchronous scoring when `approvalsReviewer` is
`"user"`, including when the reviewer changes during an active turn.
- Automatically accept ordinary `node_repl.js` execution confirmations in this
mode while continuing to surface sensitive-action checks and requests for user
input.
- Cover transitions between User approval, automatic review, and Full Access in
the app-server integration tests.
GitOrigin-RevId: 599c8ebb672728b9bf3d64c34377ddec230325f8
## 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
## What changed
- Add per-tool `analytics_result_source` requirements, with support for the
`detailed_message_search_v1` format and normal configuration precedence.
- Attach host-generated source IDs from accepted app tool results to recorded
executed tool calls for both direct and Code Mode invocations.
- Bound and deduplicate source metadata, reject caller-supplied values, and
shed optional source data before dropping tool-call or completion metadata.
- Preserve source updates across waits and retries while preventing stale or
compacted records from overwriting accepted metadata.
## Testing
Add coverage for requirement parsing and merging, trusted source snapshots,
metadata budgets, retries, waits, and compaction.
GitOrigin-RevId: 7d5e394219eb33e0ab7b6db3f039138a78569d17
## Why
Full Access already combines `approvalPolicy: "never"` with unrestricted
permissions, so confirmation-only actions do not require a model review.
## What changed
- Detect Full Access consistently across the thread and every selected
environment. Pending, failed, or restricted environments are not treated as
Full Access.
- Approve confirmation-only Guardian and MCP requests without synchronous
review, sampler prewarming, or background scoring. Cancellation, explicit
denials, and forms that require user input keep their existing behavior.
- Re-evaluate the active permission state on each turn so an existing thread
can enter or leave Full Access safely.
## Testing
Added coverage for switching approval modes, strict sensitive MCP
confirmations, failed and pending environments, and suppression of Guardian
requests and background-scoring connections.
GitOrigin-RevId: 089ab4296dde473b8e33ab8324be79c5446c46f6
## Why
`request_permissions` paths and grants need to be evaluated against the selected executor environment, including its path convention, home directory, workspace roots, and temporary directories.
## What changed
- Resolve relative and home-relative permission paths using the executor context, reject mismatched path conventions and lossy paths, and support legacy `read` and `write` path lists.
- Move grant intersection into core so requested and granted permissions use the originating environment's sandbox context. Preserve deny entries conservatively when a special path cannot be resolved.
- Keep the full originating environment with pending permission requests so delayed responses are normalized against the same context.
## Testing
- Cover POSIX, Windows, UNC, relative, and home-relative path resolution and invalid path contexts.
- Verify end-to-end app-server grants are limited to the requested workspace scope and unresolved temporary-directory denies are preserved.
GitOrigin-RevId: 730a2aacd391262e92a6314f3a5b6c262e3dca10
## 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
## 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
## What changed
- Preserve the selected app account `link_id` when building native MCP tool
approval requests.
- Add `link_id` and `link_is_implicit` to the elicitation metadata, recognizing
IDs with the reserved `implicit_link::` prefix as implicit links.
## Testing
- Extend the app link policy integration test to verify elicitation metadata
for default, explicitly selected, and implicit account links.
GitOrigin-RevId: a76731d9f5f3640318187511ccdc0dc82967a9c5
## Why
Remembering an app tool approval by connector and tool alone could reuse an
approval when the same tool was called with a different selected account.
## What changed
Include the app `link_id` in MCP tool approval keys so remembered session
approvals apply only to the account link that was approved. Calls for another
link, or calls without a link selector, request their own approval.
## Testing
Add an integration test covering calls across two link IDs and the legacy
no-selector case.
GitOrigin-RevId: 7dd88330b75de312469020d5892dfb2fa5d5fada
## Why
Git root discovery is optional metadata work, but filesystem probes can block. They should not exhaust Tokio's blocking pool, delay runtime shutdown, or prevent later turns from observing repository changes.
## What changed
- Add a shared `GitRootDiscovery` service that coalesces concurrent lookups for the same working directory and limits probes across directories.
- Run probes on detached threads, retain in-flight work across caller cancellation, and discard completed results instead of caching them.
- Use the service for turn and memory metadata enrichment, abort unused turn enrichment when its state is dropped, and limit memory metadata waits to one second.
## Testing
Add coverage for probe sharing, capacity limits, cancellation, fresh discovery, runtime shutdown, memory timeouts, and repositories restored after startup prewarming.
GitOrigin-RevId: bca46fc263e7a12a2f69146d8a0b3e7c7e0846cb
## What changed
- Add the under-development `mcp_oauth_refresh_coordination` feature and pin its selected refresh mode to each MCP connection.
- Treat the refresh mode as part of OAuth connection identity so runtime and MCP configuration refreshes reconnect when the mode changes.
- Keep legacy refresh behavior in builds where coordination is unavailable and emit a warning when coordinated mode is selected.
## Testing
- Cover connection replacement when the feature is toggled through both runtime and MCP configuration refresh paths.
GitOrigin-RevId: 89596150f0338e6ede0e712badecea4450831766
## What changed
- Add `approvalsReviewer` to the experimental `turn/settings/update` API.
- Apply reviewer changes to subsequently captured steps and new background approval requests while preserving pending approvals and future-thread defaults.
- Allow reviewer-only updates without `step_model_switching`, while continuing to enforce managed reviewer restrictions and model-required auto review.
- Use an explicit live reviewer update for MCP approvals while retaining refreshed thread defaults for clients that have not set one.
## Testing
- Cover switching between user and automatic review during a live turn, future-turn isolation, and managed-authority rejection.
GitOrigin-RevId: e1216b09c6a7972c2c9eaf4955d8d57d60604ca1
## Why
Unified exec can run commands on an OS that differs from the Codex host. Using
host path and shell semantics for approval checks can therefore miss dangerous
Windows and PowerShell commands sent to a remote executor.
## What changed
- Thread the executor platform through command parsing, executable-name
normalization, and dangerous-command classification. Legacy executors fall
back to the host platform, while unknown reported platforms use Windows rules
conservatively.
- Bound model-facing `exec_command` rejection messages and avoid echoing the
rejected command into the error.
- Keep deterministic process IDs reserved after release so rejected test
commands cannot reuse an earlier ID.
## Testing
- Add coverage that a long dangerous PowerShell command targeting a remote
Windows executor is rejected with a bounded response.
- Verify deterministic process IDs advance after release.
GitOrigin-RevId: a26b4a63ea3d6df7140032b88b3e7aec0b85b948
## What changed
- Centralize the allowlist for bundled MCP cleanup hooks and use it for both local and executor-discovered plugins, including `unified-computer-use` cleanup through `cua_repl`.
- Mark matching cleanup hooks as trusted built-ins so they run without saved hook trust and remain active when regular hooks or their per-hook state are disabled. Plugin enablement and managed-only policy still apply.
- Hide built-in cleanup hooks from hook listings and lifecycle notifications while retaining their metrics. Keep the built-in classification out of serialized protocol data.
## Testing
- Cover allowlist boundaries, trust and enablement behavior, inline and file-based hook declarations, MCP success and error responses, hook listing, lifecycle notifications, metrics, and protocol serialization.
GitOrigin-RevId: f93b7bc99f4ed9694f529def8ec383b45f31430e
## Why
When the remote Sites plugin is installed and its local bundle is available, exposing the bundled copy as well creates duplicate catalog entries and can load the wrong skills.
## What changed
- Persist an account- and backend-scoped exclusion for `sites@openai-bundled` once the remote replacement is installed and loadable.
- Apply the exclusion to plugin catalogs, search, direct reads and installs, runtime loading, hooks, and skill caches while preserving the remote plugin's server-owned enabled state.
- Restore the bundled fallback when the remote plugin is absent or its local files are unavailable, and throttle repeated availability checks for 60 seconds.
## Testing
- Added manager tests for persistence across restarts, account/backend isolation, fallback restoration, concurrent checks, and refresh throttling.
- Added app-server and agent-turn tests confirming that catalog, search, and skill loading prefer remote Sites.
GitOrigin-RevId: bc1154f79d3107910fee9fa27389a0ddb48ce6cf
## Why
Custom MCP tool names may contain user data, so they should not be added to Guardian metric tags.
## What changed
- Add the sanitized `tool` tag to Guardian review counts for MCP calls served by the OpenAI Apps server.
- Keep custom MCP calls and Guardian duration and token-usage histograms on their existing tags.
## Testing
- Cover network access, OpenAI app tool calls, and custom MCP tool calls in the Guardian metrics test.
GitOrigin-RevId: 6bd09ae461eab504f8f07bf7d0c9d1c0aa1317c9
## What changed
- Add `SectionRegistry::compose` to group collected authorization fragments and transcript entries while preserving fragment boundaries and registration order.
- Use the shared composed context in the core Guardian prompt and Guardian V2 synchronous and asynchronous reviewers.
- Keep transcript rendering, retention, and delta handling with each caller while removing duplicated section grouping and authorization rendering.
## Testing
- Cover composition for synchronous and asynchronous targets, including root conversation role handling, trusted user answers, transcript entries, empty context, and contributor failures.
- Verify the synchronous reviewer preserves authorization fragment boundaries in prompt items.
GitOrigin-RevId: 9268343631fd6417bb05a3c0abce883559a37685
## What changed
- Retain MCP invocation metadata for Node-backed REPL servers so nested
elicitations can resolve the enclosing tool call.
- Use a valid, same-server `callId` from the current turn as the Guardian
assessment target while continuing to review each nested action and connector
independently.
- Fall back to an elicitation-specific target when `callId` is missing,
malformed, unknown, from another server, or from an earlier turn.
## Testing
- Cover ordinary and strict reviews for `node_repl` and `cua_repl`, multiple
nested decisions, unrelated servers, and invalid or stale call IDs.
GitOrigin-RevId: adea4b5cff96726a540ff3b3ea902b0f5dd99299
