## What changed
Register `fs/writeBlock` and require filesystem initialization before handling it. Validate the file handle ID, then return an invalid-request error stating that writable file streams are unsupported.
GitOrigin-RevId: 13440e497a88ca6505ee60d95bbb771d3cae06b8
## Why
Without a UI flush tick, expired paste detection can cause Enter to be treated as part of a paste instead of submitting buffered input.
## What changed
Flush due characters and paste buffers before classifying Enter during submission, including in Vim insert mode. Preserve the buffered text when submission is blocked for a parent-owned thread.
## Testing
Add regression tests for held characters and buffered bursts with and without Vim mode and parent-owned threads, plus a snapshot test for Enter submitting expired input in Vim insert mode.
GitOrigin-RevId: 1d0d309b57e9d2151e0d9c1774a3401d05413d3c
## Why
Explicit launch permissions such as `--yolo` can be lost during reconnect recovery, causing subsequent input to use different approval and sandbox settings.
## What changed
Remember explicit local launch approval policies, approvals reviewers, and permission profiles at startup and when creating new sessions. Restore the approvals reviewer independently of profile overrides during reconnect recovery. Keep constrained profiles that cannot be expressed through legacy turn parameters under server control.
## Testing
Add regression coverage for `--yolo` across new sessions and reconnects, later permission-picker choices, reviewer-only overrides, and server-owned constrained profiles. Extend the reconnect integration test to submit the preserved draft and verify its approval and sandbox settings.
GitOrigin-RevId: 7a77dc41b2a357ab17fec9ba6cb2361f4a8f4aa5
## What changed
- Mark `api_key_model_discovery` as stable and enable it by default, while preserving explicit user opt-outs and provider catalog requirements.
- Fall back to bundled models when discovery is disabled at runtime, including when a remote catalog has already been loaded.
## Testing
Extend model-list coverage for default enablement, app overrides, and user configuration precedence. Add a regression assertion that runtime opt-outs hide loaded catalogs without additional fetches, and update daemon compatibility tests for the new default.
GitOrigin-RevId: feaf732757792cc58063f21a5327d0e633f079f4
## Why
New error variants should not prevent older clients from handling app-server errors.
## What changed
- Deserialize unknown `CodexErrorInfo` strings and objects as `Other` in Rust, while preserving known variants and the `"other"` serialization.
- Allow arbitrary strings and objects in the exported JSON schemas and TypeScript type.
- Update Python SDK generation and generated models to accept unknown error strings and object payloads.
## Testing
Add Rust tests for `"other"` round trips, unknown string and object fallbacks, known structured variants, and rejection of null, boolean, numeric, and array values.
GitOrigin-RevId: 40c17896019924b2ef7970ee2205ea0154d0d1b3
## Why
Older executors can ignore the file open mode and silently return a read-only handle for a writable open.
## What changed
Add `ExecServerClient::fs_write_block` and require the `file_write_streaming` capability for both block writes and `FsOpenMode::Replace` opens. Return a protocol error before sending either request when the executor lacks support, while allowing read-only opens.
## Testing
Add a WebSocket regression test verifying that unsupported writable opens and block writes are rejected locally and read-only opens still reach the executor.
GitOrigin-RevId: f206e1da80c12af76ead14c75f8ffe71f82cce89
## What changed
Render Control, Shift, and Alt as `⌃`, `⇧`, and `⌥` on macOS, and Control as `^` on Linux. Keep textual modifier labels on other platforms.
Route hard-coded shortcut hints through the shared key-label formatter so footers, overlays, pickers, and transcript notices use consistent notation.
## Testing
Update modifier-combination and chord label assertions, along with rendering snapshots and narrow-layout expectations for the shorter hints.
GitOrigin-RevId: 0d20fa218e4a11dfe7a81c8de64f8d25e093af03
## What changed
Pin `dtolnay/rust-toolchain` to a revision from `master` in the argument-comment lint package CI and release build workflows. Both continue to request `nightly-2026-08-20`.
GitOrigin-RevId: 63801f669b91d4bfa1d69fc05d728d838ca8add0
## Why
A saved side conversation transcript can outlive its ephemeral server thread. The resulting `turn/interrupt` error prevented side conversation cleanup and blocked archiving the parent thread.
## What changed
Treat the specific `thread not found` interrupt error as successful interruption for replay-only side conversations, allowing unsubscribe and local cleanup to finish. Continue sending interrupt requests because replay-only conversations may still be running after reconnect.
## Testing
Expand archive lifecycle coverage for local daemon and remote server connections to include live side conversations, replay-only conversations with existing threads, and replay-only conversations whose threads are gone. Assert that interruption, unsubscribe, archiving, and local cleanup occur.
GitOrigin-RevId: 0528c9f5b5278931fd3a7941073071117f2f3612
## Why
The TUI unconditionally forwarded its resolved `web_search` mode, allowing implicit client settings to replace server defaults or saved thread settings.
## What changed
- Forward `web_search` on thread start, resume, and fork only when the winning choice comes from session flags or a selected profile.
- Translate legacy search feature flags into a canonical `web_search` override while preserving precedence, and remove those flags from forwarded feature overrides.
- Honor shared feature requirements for embedded sessions while letting remote servers enforce their own requirements on explicit search choices.
## Testing
Add coverage for configuration origins, canonical and legacy search precedence, and embedded versus remote feature requirements. Extend the Responses integration test to verify that a remote server's live search setting survives thread start and resume when the client's configuration disables search.
GitOrigin-RevId: c714f9cf0321aca9a5fa292a350720af5a2d02d4
## What changed
Return `Arc<EnvironmentInfo>` from `ExecServerClient::environment_info()` so repeated reads and cloned clients share cached metadata without cloning its contents. Wrap metadata from initialization and lazy RPC fetches in `Arc`, while preserving the owned return value in `LazyRemoteExecServerClient`.
## Testing
Extend the cache test to assert pointer identity across cloned clients for both initialization-provided metadata and lazy fetching after connecting to a legacy server.
GitOrigin-RevId: 3f56f4d51a4aebc0c4821c0b128a70efb23933d4
## Why
Guardian review continuations can repeat retained-instruction omission notices even when availability has not changed. When retained context becomes complete, earlier notices also need to be superseded explicitly.
## What changed
Track delivered user-instruction and assistant-context omission states in host metadata scoped to the retained-instructions section. Suppress redundant notice-only sections when the state is unchanged, and emit an availability-change notice when an earlier omission clears, including when no retained originals remain.
Treat missing delivery metadata as unknown. Matching notice text in coalesced tool output cannot establish host delivery, and incomplete history entries discard omission delivery proof.
## Testing
Add coverage for repeated notices, independent histories, availability recovery, and matching notice text in multimodal REPL output. Extend app-server review-continuation tests to assert that each effective classifier context contains the retained user-instruction omission notice exactly once.
GitOrigin-RevId: c6d52dd1dafb284af34b46174aecc7c8cc9b8404
## Why
Retained Guardian classifier conversations already contain previously delivered synchronous reviews. Appending those reviews again repeats evidence and consumes input budget.
## What changed
Track each review completion with a stable identity in host-only history metadata. Filter already delivered reviews before composing and budgeting continuation input, while preserving distinct completions even when their text is identical.
Fresh conversations receive the review evidence again, and shortened messages no longer count as complete delivery. Delivery metadata stays out of model requests.
## Testing
Add coverage for delivering retained reviews once, appending newly completed reviews, restoring evidence after failed streams, and retaining reviews when history lacks complete delivery metadata.
GitOrigin-RevId: b8d58ffdd7d9c484676af707c65fa50020aae73c
## What changed
- Add `features.guardianv2.async_classifier_mode` with `snapshot` and `conversation` modes. Local configuration overrides model defaults; snapshots remain the fallback.
- Retain completed classifier requests and responses in conversation mode, appending transcript deltas without repeating previously delivered user instructions.
- Rebuild retained history when authorization, classifier instructions, or parent compaction changes, after an incomplete response, or when the next request exceeds the configured token limit. Add `async_classifier_conversation_token_limit`, defaulting to 100,000 tokens, while preserving the model's input limit for fresh requests.
- Serialize conversation requests in observation order, bound outstanding work, and reject stale work after authorization changes or thread shutdown. Record retained and new input token costs.
## Testing
Add tests for configuration precedence, history reuse and resets, ordered admission, queue overflow, and stale or stopped requests. Exercise the approval matrix in both modes, including resume, fork, mandatory review, authorization changes, and stream failure after an early score.
GitOrigin-RevId: 35b9800104648cde4f467086e678dee12458c3e5
## What changed
- Add request preparation that preserves committed classifier history and appends transcript deltas, filtering instructions and images already delivered. Account for the complete input budget before transferring history; rejected requests leave the checkpoint and cursor intact.
- Add retained sampling that publishes the early score while collecting completed assistant output and encrypted reasoning for reuse. Preserve item IDs and input metadata, and discard unusable history without retracting an already published score.
- Keep the existing snapshot sampling path and share input preparation and execution with the new retained path. Retained sampling is not yet wired into the active backend.
## Testing
Add coverage for transcript and image deduplication, immutable history prefixes, budget boundaries, rejected preparation, WebSocket output retention, and retention failures that preserve the early score.
GitOrigin-RevId: 0fa33db8b8ea4483eb64d75e207690f1d349123b
## What changed
Remove the repository-root `AGENTS.md` from `codex-rs/core` test data and from the root Bazel `exports_files` list.
GitOrigin-RevId: 644052d50f79b4c798baf40e53aab824f0fd73c3
## What changed
When the model catalog is provided in context and model overrides are exposed, tell V2 `spawn_agent` callers to inherit the current model unless the user explicitly asks for a different model. Include this guidance even when spawn metadata is hidden.
## Testing
Extend tool-description tests to cover hidden metadata with both inline and context catalogs, and update request-level assertions for V1 and V2 model override guidance.
GitOrigin-RevId: b70b0f09b0b5987a0eccba578196b8ef7cee4baf
## Why
Naming a paginated thread before its first turn must make it durable so it can be resumed immediately and after an app-server restart.
## What changed
Persist the live rollout before updating name metadata in SQLite. Propagate persistence failures before saving the name.
## Testing
Add thread-store tests for empty-thread persistence and failure handling, plus an app-server regression test that resumes a named empty thread before and after a restart and completes its first turn.
GitOrigin-RevId: 2f7556c7de65f6a86177130c39e1a17951496931
## What changed
Register `browser_annotation_api` as a stable, default-enabled feature intended for requirements rather than user configuration. It controls websites opening and customizing annotation tools in desktop apps, independently of ordinary user-driven annotation.
Add the key to the configuration schema, extending the browser requirements support introduced in https://github.com/openai/codex/pull/18956.
## Testing
Extend `browser_feature_requirements_are_valid` to verify that enterprise requirements setting `browser_annotation_api = false` disable the feature.
GitOrigin-RevId: 8c7bb65932e08560040958b4472fe45f6d1c916f
## Why
Switching to a fork can interrupt approval routing for source and child threads that remain on a local daemon or remote app server. Their pending requests and event listeners need to survive the switch.
## What changed
- Retain background request routing and event listeners for tracked threads when switching to a daemon-backed fork, preserving pending approvals without duplicating them.
- Register the source thread before the `fork_thread` tool creates a fork so its requests remain routable during the operation.
- Route requests for the currently viewed thread to the active UI, even when that thread is also tracked in the agents overview.
## Testing
Add regression coverage for local daemon and remote forks retaining source and child threads, replaying pending requests when returning to the source, and displaying active-thread file-change approvals. Verify that embedded forks still discard the source thread state and that `fork_thread` registers its source before issuing the fork request.
GitOrigin-RevId: 63139e2b9a6eff373ab05a31e0140a432ce47389
## Why
Failed shell snapshot captures could leave processes launched by shell startup files running after the capture ended.
## What changed
Run each Unix shell snapshot capture in its own process group and kill that group on failure, timeout, or cancellation. Keep the capture leader unreaped until output and exit status are validated so its process group ID cannot be reused during cleanup. Preserve startup helpers after a successful capture.
## Testing
Add regression coverage for startup child cleanup after nonzero exits, invalid output, cancellation, and timeout, alongside fallback and retry behavior. Verify that successful captures preserve delayed startup output and background services and reuse the cached snapshot.
GitOrigin-RevId: a56ef5d8615325952bdc8bb78727de1559e7e9c7
## 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
- Add `read` and `replace` modes to `fs/open`, defaulting omitted modes to `read` for existing callers.
- Define `fs/writeBlock` request and response types for offset-based writes, with a maximum decoded chunk size of 1 MiB.
- Add the `fileWriteStreaming` capability, defaulting to `false`. Keep writable streams unsupported in the server and explicitly reject replacement opens.
## Testing
Add a protocol test verifying that legacy `fs/open` requests without a mode deserialize as read-only opens.
GitOrigin-RevId: 4d3997717e821cd1f62df12a03c8eec36332a66d
## What changed
Fetch optional security setup notices asynchronously for local ChatGPT sessions using the OpenAI provider. Verify that saved credentials match the connected app server, ignore stale account responses, and validate notice text and HTTPS actions on `chatgpt.com`.
Show notices as dismissible inline banners with a browser action, while giving applicable usage banners priority. Preserve notice state and dismissal across thread changes, reconnects, and banner replacement for the same account and user.
## Testing
Add authenticated HTTP fixture tests, notice validation tests, and TUI tests covering stale responses, dismissal persistence, usage banner priority, browser actions, and wrapping in narrow terminals.
GitOrigin-RevId: b8635fcf6a26b42b6a06640601c8517d8d5e47e8
## What changed
Allow API-key sessions using the OpenAI provider to forward explicit cyber access programs when `features.api_key_cyber_access_programs` is enabled, regardless of `features.api_key_model_discovery`.
## Testing
Expand the app-server test to cover all four combinations of the two feature flags, checking forwarded programs and failures when forwarding is disabled. Default the API-key test helper to model discovery disabled.
GitOrigin-RevId: f9cf1d2bd7c231645d50178cc6721b55a8d63268
## What changed
Delete the root `AGENTS.md`, the skills under `.codex/skills/` and their supporting scripts, tests, and references, and `.codex/environments/environment.toml`.
GitOrigin-RevId: 555482804c79d17f9aa5587c4d26ea03337db8d9
## Why
Token-budget truncation uses byte counts to estimate tokens, but previously scanned the entire string to count characters, including the discarded middle.
## What changed
Use UTF-8 boundary lookups to select the retained prefix and suffix for token-budget truncation. Skip character counting for a zero-token budget and defer it to paths that need character counts.
## Testing
Add regression coverage for multibyte characters crossing both retained boundaries, checking the truncated output and original token estimate.
GitOrigin-RevId: 37fee0c8d4ede8e74b491d3d641b0ea0e9003dd4
## What changed
Preserve the underlying error in `LocalStateDbStartupError` and use its error chain to decide whether CLI startup failures qualify for automatic database backup and recovery.
Replace corruption checks against error-message text with SQLite result-code classification, including SQLx extended result codes and directly wrapped `SqliteError` values.
## Testing
Update the embedded startup corruption test to verify that the startup error retains a detectable SQLite corruption cause and the failed database path.
GitOrigin-RevId: 6e8ecd0ee67f3c239183ddea3fc75c4e4cd5744a
## Why
Session index updates perform synchronous file I/O and wait on a blocking mutex inside async functions, blocking runtime threads.
## What changed
Run appends and thread-name removals in `spawn_blocking` workers. Acquire a Tokio mutex before dispatch and keep its owned guard in the worker so queued and running updates remain serialized even if the awaiting task is cancelled. Move the lookup's file-existence check into its blocking worker as well.
## Testing
Add tests for lock retention after cancellation of running and queued updates, write ordering after cancellation, and removal that preserves other thread names and malformed lines.
GitOrigin-RevId: 368745d8e5bbf24c527facc0a62ddd0fb76f0ffb
## What changed
- Move the lint to `nightly-2026-08-20` and Dylint `6.1.0`, updating CI, release artifacts, Bazel configuration, and installation instructions.
- Adapt to the updated rustc and Clippy APIs, including resolving `Option::None` through type aliases when checking for uncommented arguments.
- Set the `chatgpt` crate's `recursion_limit` to `256`.
## Testing
Extend the UI fixture to expect an uncommented-argument warning for `OptionalString::None`, where `OptionalString` aliases `Option<String>`.
GitOrigin-RevId: 34e1b173972f0bfa9b6c8fd3342ca652539efe2f
## Why
Publishing an older alpha or alpha hotfix can overwrite npm dist-tags that already point to a newer release.
## What changed
Check each package's current npm dist-tag before publishing an alpha release, using the existing release version comparison and workflow concurrency lock. Handle `alpha` and platform-specific tags independently for Codex, the SDK, and the responses API proxy.
When the release does not advance the current version, publish under `release-<version>-<tag>` to keep it installable without changing the alpha or `latest` tags. Abort publishing if the registry lookup fails.
## Testing
Add unit tests for package and platform tag selection, older releases, dotted alpha hotfixes, missing tags, and registry lookup failures.
GitOrigin-RevId: 9c0edec1149bf78c01aac0f700e963630316eaea
## What changed
Rename `file_read` to `file_handle` and `FileReadHandleManager` to `FileHandleManager`. Update the filesystem handler field, handle ID validation helper, and related constants to use file handle terminology. File operations, limits, and error messages remain unchanged.
GitOrigin-RevId: 25ad81a917386da23fd42ecff79bd47a482ce430
## Why
A database can open successfully while containing corruption that startup previously missed. Detecting it before runtime initialization completes allows recoverable databases to be rebuilt while preserving the damaged files.
## What changed
- Run `PRAGMA quick_check(1)` when opening writable databases, with a 100 ms scan budget. Share attempts across cloned `SqliteConfig` instances by file identity so repeated opens skip validation and replacement files are checked again.
- Back up and rebuild runtime databases that support recovery. Report corruption without rebuilding thread history or other databases without an automatic recovery policy. Treat interrupted or locked checks as incomplete rather than confirmed corruption.
- Collect startup backups across recovery retries and include their locations in the app server warning. Explain that saved conversations can restore the thread list and history, while some database-only metadata may be unavailable.
- Record confirmed corruption in `codex.sqlite.corruption.count`.
## Testing
Add tests for recovery policies, backup preservation, validation caching and concurrency, cancelled and interrupted checks, and locked databases. Add app server coverage for recovery warnings and saved conversation restoration, a TUI warning snapshot, and message board post preservation.
GitOrigin-RevId: 43e26328888e0301def642d69bd324acee867566
## What changed
Run direct `read_file` operations in a blocking task with bounded read chunks and cancellation checks. Dropping the async caller cancels further reads between chunks. Preserve the 512 MiB size limit, symlink-following options, and regular-file validation.
## Testing
Add regression coverage for 4 MiB reads and oversized-file rejection with both symlink-following options, cancellation after the first chunk, and FIFO rejection without waiting for a writer.
GitOrigin-RevId: 19f2816de15a03bd89b463de6b002367e6a692a9
## What changed
Move directory enumeration and candidate metadata lookups into blocking workers, batching updated-time candidate collection into one worker per scan. Check cancellation between entries and metadata lookups so dropping the async waiter stops further scanning.
Share synchronous rollout metadata resolution with the async helper, preserving plain-file precedence over compressed siblings. Retain scan limits, stable file ordering, and directory-level early stopping for creation-time listings.
## Testing
Add tests for cancellation during directory enumeration and candidate discovery in both flat and date-partitioned layouts, including plain-sibling metadata precedence and filtering of unrelated files, directories, and symlinks.
GitOrigin-RevId: 324289003b29a7d2b3220da03e3c002ab5e23eec
## What changed
Run rollout file opening, metadata lookup, seeking, reading, and record
projection together in a single `tokio::task::spawn_blocking` call. Use
synchronous file I/O in `read_projection_steps_sync` and carry the current
tracing span into the blocking task.
GitOrigin-RevId: a231a8b9a328f6d2e2095a29cde1718ccb6b3901
## Why
Compressed snippet searches schedule a blocking task for every line read. Keeping the scan on one worker avoids repeated task scheduling.
## What changed
Use `RolloutLineReader::find_map` to scan compressed rollouts on a single blocking worker, stopping at the first matching snippet. Preserve read metrics and stop delivering records when the search is canceled, including cancellation during a read.
## Testing
Add tests for first-match snippets, case-insensitive literal matching, malformed records, missing matches, stream errors before a match, and cancellation between lines or during a read.
GitOrigin-RevId: 295206b1a00e861bc44606bd843f6ed8c3e6aae1
## Why
An inaccessible user profile directory can prevent sandboxed PowerShell from preserving its working directory and resolving relative paths in a project beneath that profile.
## What changed
Pass `USERPROFILE` to elevated setup and refresh. When the command's working directory is beneath an absolute profile path, grant the sandbox group non-inheriting `FILE_READ_ATTRIBUTES` access to the profile directory.
Preserve existing permissions and deny ACEs during non-inheriting ACL updates without propagating changes to descendants. Expand generic deny masks when checking for conflicting permissions.
## Testing
Add a Windows regression test that checks PowerShell's working directory and reading `.\public.txt` beneath a protected profile. Add an ACL test that verifies a non-inheriting grant leaves an unprotected child's security descriptor unchanged.
GitOrigin-RevId: 7f142a7ffdee795c3255d4a6e79568fa9391c055
## 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
## What changed
Register `in_app_voice` as a stable, default-enabled feature and add it to the configuration schema. The gate is intended for managed requirements to control permission for voice in desktop apps; permission does not establish voice availability or provider support.
## Testing
Add app-server coverage for reporting `in_app_voice = false` from `requirements.toml` in both requirements and feature-list responses. Extend the CLI feature-list test to verify that cloud-managed requirements disable the gate despite a user-configured `true` value, without rewriting the user configuration.
GitOrigin-RevId: 6ec3bc0176fdb3e1714b0136bf812bd56591f6a1
## Why
When a turn ends, recovered question answers are appended to the composer draft. If that draft starts with a shell or slash command, submitting the combined text can interpret it as a command instead of a prompt.
## What changed
Escape command prefixes before appending recovered answers so the combined draft stays editable and submits as literal prompt text. Convert shell mode's separate `!` prefix into escaped composer text, and leave slash-prefixed paths such as `/tmp/file` unchanged.
Keep prefix escaping and answer recovery in the same Vim undo edit.
## Testing
Add regression coverage for immediate and queued submission, whitespace around command prefixes, slash-prefixed paths, and already escaped commands. Extend Vim undo coverage to shell and slash command drafts, including shell mode, and add a snapshot of the recovered shell draft.
GitOrigin-RevId: 03c40f72a834a65bf367103cb9170a86dc35ad98
## Why
Gateways that inspect request bodies incrementally need routing fields before potentially multi-megabyte prompts so they can route requests without buffering the input first.
## What changed
Serialize `model`, `stream`, and optional `service_tier` before `input` in both HTTP Responses requests and WebSocket `response.create` messages. Preserve field values and omission of unset `service_tier`.
## Testing
Add an HTTP regression test with a 2 MiB input that checks routing field order and unchanged JSON contents. Extend WebSocket serialization coverage to assert the routing fields appear first, and check that HTTP requests omit unset `service_tier`.
GitOrigin-RevId: dbc174e1fe7ca30d0c00f075bc37dd5f97b27b37
## Why
`windows.allowed_sandbox_implementations` permits MXC regardless of the listed implementations. Managed requirements need a separate way to block both explicit MXC configuration and automatic selection through `prefer_mxc`.
## What changed
Add the optional managed requirement `windows.allow_mxc`. Setting it to `false` disables automatic MXC selection and rejects explicit `windows.sandbox = "mxc"` configuration with an error identifying the requirement. Leaving it unset or setting it to `true` preserves existing behavior.
## Testing
Add coverage for the requirement alongside allowed sandbox implementations, preference resolution, and cloud-managed opt-out behavior for implicit, elevated, and explicit MXC configurations.
GitOrigin-RevId: a458a7b4c92cf17e687dbbad8208884d1c097da5
## Why
Session commands with `--remote` should use the selected server's authentication and execution environment, even when the caller has workload identity configured and cannot read its identity token.
## What changed
Connect `queue`, `archive`, `unarchive`, and `delete` directly to the explicit remote app server after validating configuration. Skip caller credential and runtime initialization, use remote thread parameters, and preserve the caller's working-directory override.
## Testing
Add end-to-end coverage with an unreadable caller identity token to verify queued turns use the server's API key and remote archive, unarchive, and deletion operate on the selected server. Add configuration-validation coverage for all four commands and `--no-daemon` rejection coverage for remote unarchive and delete.
GitOrigin-RevId: 5da7abbb40506cd754a8863e1a9d8fad9b1f09a9
## 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 and parent requests can arrive in either order after a command yields. A shared response sequence can let parent requests consume Guardian review decisions.
## What changed
Match Guardian responses separately and drive parent responses from the latest tool output in the strict automatic network review test. Poll running sessions with `write_stdin` until the process exits, and assert that the command and network request each receive Guardian review.
GitOrigin-RevId: 21300abd65a3b2d439e8c1ea04e4429d5bd528dd
## 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
Adam Perry @ OpenAI