## Why
Windows sandbox setup and permission choices need to reflect the connected app server's requirements and the active thread's executors.
## What changed
- Refresh sandbox configuration and requirements for the current working directory when threads or directories change, before setup, and after setup completes.
- Use the thread's observed environments to determine whether sandbox setup is local, remote, mixed, or unknown.
- Defer initial prompt submission while requirements load and preserve the draft if loading fails or required setup is unavailable.
- Hide sandbox setup choices and elevation commands that requirements or executor selection disallow.
## Testing
Add coverage for observed thread hosts, app-server configuration reads, draft preservation and recovery after read failures, deferred initial prompts, and non-admin-only setup choices.
GitOrigin-RevId: d85ac66d18951b752e71d1a6bf25404eeedcad02
## Why
Daemon recovery reloads threads but leaves interrupted work unfinished. Resume eligible work automatically from the saved conversation, even without a connected client.
## What changed
- Attempt one new continuation turn immediately after restoration, including in Plan mode. Mark the old turn interrupted and supply recovery context without creating a user message or granting new authorization.
- Require an idle thread, matching permissions, and the same single local environment configured by the thread. Skip completed, aborted, or superseded work and snapshots without environment identity.
- Add `continue_turn_if_idle` with an atomic previous-turn check so newer tasks or standalone settings changes invalidate pending continuation.
- Preserve the output schema, service tier, and root turn ID, and emit a “Resuming interrupted work” warning when continuation starts.
## Testing
Add daemon restart coverage for continuation without a client, Plan mode, permission and environment mismatches, and legacy snapshots. Add core coverage for preserved continuation metadata, absence of user-message events, rejection of superseded continuations, and exclusion of remote execution from recovery snapshots.
GitOrigin-RevId: 2b290b75d9399fc58258bcda2e85c74e9b5b3b09
## What changed
Add `codex-mermaid`, a standalone crate that renders supported subsets of flowchart, sequence, state, class, and ER diagrams as Unicode text. Expose `render` for plain text and `render_spans` for semantic node, edge, and text spans that callers can style.
Enforce source, diagram, canvas, and display-width limits. Return errors for unsupported input or exceeded limits without producing partial diagrams, leaving source fallback to callers. Include documentation and a stdin rendering example.
## Testing
Add snapshots for each diagram family, checks for relationship endpoints, Unicode labels, semantic spans, truncated input, and size limits, plus edge reconstruction for all 512 directed three-node graphs in all four layout directions.
GitOrigin-RevId: 5b4e65d9152abce5d833e4e6b12113a5a21f700e
## What changed
- Add `WindowsSandboxConfig` helpers to read app-server configuration and managed requirements, preserve legacy feature flag fallbacks, and select an allowed setup mode with elevated mode preferred when the configured mode is disallowed.
- Store the sandbox host classification in thread session state using reported environments across start, resume, fork, and replay paths. Classify missing or empty environments as `Unknown`, and distinguish local, remote, and mixed selections.
## Testing
Add coverage for managed setup restrictions, legacy configuration precedence, unknown policy state, environment classification, and remote host state in inactive-thread replay.
GitOrigin-RevId: ae8920d9c2a1c80b641ac90617ae607b171bd55a
## What changed
- Add an optional `WorkspaceRoutingResolver` to `AuthManager` and pass session configuration into routing lookups. Callers without a resolver retain existing routing behavior.
- Apply resolved backend origins and account routing overrides to eligible ChatGPT Responses HTTP requests and WebSocket handshakes. Preserve endpoint paths, validate routing values, and reject HTTP redirects for routed requests, including `NO_CONSTRAINT` routes.
- Key cached WebSocket connections by destination, routing header, and auth revision. Rebuild request setup after credential refreshes and reject account changes during setup or routing discovery.
- Serialize routing lookups per session and retain whether the session has previously been routed.
## Testing
Add tests for path preservation, provider exclusions, concurrent routing lookups, unavailable resolvers, workspace mismatches, redirect rejection, and credential refresh or account changes during request setup.
GitOrigin-RevId: 2b6609019e84a315745c94858955f3f16426fe7f
## Why
WSL interop can block while launching the Windows terminal probe, stalling TUI startup. Inconclusive detection can also enable keyboard enhancements that break dead-key composition in VS Code on WSL.
## What changed
- Give the Windows `TERM_PROGRAM` probe a one-second deadline, with launch and cleanup handled on a worker thread. Cache the result, including timeouts, and kill and reap probes that outlive the deadline once launch returns.
- Distinguish VS Code, other terminals, and unknown detection results. Disable keyboard enhancements on WSL for VS Code or unknown results, while preserving the `CODEX_TUI_DISABLE_KEYBOARD_ENHANCEMENT` override.
## Testing
Add regression tests for probe output and exit status, failed launches, timeouts during launch and execution, and child cleanup. Extend detection and override tests to cover unknown results.
GitOrigin-RevId: 39e980b0d7cf3d6716ee61597d6002b0fd59651e
## What changed
- Accept and ignore `features.personality` in user configuration, profiles, and managed requirements.
- Remove feature gating from `personality = "none"`, which strips the literal `# Personality` section when preparing model catalog instructions. Preserve explicit base instructions and existing thread instructions.
- Document deprecated personality fields in configuration, app-server schemas, and the Python SDK: `friendly` and `pragmatic` no longer select a style, and `supportsPersonality` is always `false`.
## Testing
Add regression coverage for ignored legacy flag values and managed requirements, personality opt-out behavior with the flag absent or set to either value, role overrides, and preservation of explicit base instructions, including empty strings.
GitOrigin-RevId: 4e12c66b42bfc59d6f151a5b3c28fadc6654ae99
## Why
Managed daemon recovery snapshots previously saved only loaded thread IDs, without identifying active turns or preserving their turn-specific options.
## What changed
- Capture regular, uncanceled turns after their input is recorded, saving the turn ID, output schema, service tier, and cyber access program alongside persisted thread IDs.
- Store interruption metadata atomically in the existing candidate array format so older servers can still read thread candidates.
- Begin snapshotting once admitted operations drain, while turns may still be running. Run snapshot collection and thread listener attachment independently of the event loop to keep forced shutdown responsive.
## Testing
Add coverage for running, completed, canceled, and compacting turns; recovery readiness for automatic and user turns; admitted resumes during shutdown; forced shutdown with a blocked rollout writer and child listener attachment; and legacy candidate-array compatibility.
GitOrigin-RevId: ed46342c3a5c71b09c48fa9acece2f15ae748e2f
## What changed
Reject `request_plugin_install` calls from non-root agents with an error returned to the model before parsing arguments or prompting for installation. Apply the restriction to both legacy install requests and recommended plugin requests.
## Testing
Add unit and integration tests for both request formats, verifying that subagent calls return the root-only error and emit no installation elicitation request.
GitOrigin-RevId: f23bbc6dd34d69366b81e4033bd554bba1fb816a
## Why
Clients need widget presentation details to render MCP Apps without waiting for the full MCP catalog, including when replaying saved history.
## What changed
- Add `mcpToolCall.mcpAppUi` with the invoked descriptor's `resourceUri` and `preferredModelDisplayMode`, and carry it through tool-call events and saved history.
- Support `fullscreen` and default to `inline` when the display preference is missing or unsupported.
- Keep existing resource URI fields for compatibility. Leave `mcpAppUi` null for older history and tools that declare widgets only in result metadata, where clients still use catalog discovery.
- Update protocol schemas, TypeScript and Python types, and app-server documentation.
## Testing
Add parameterized integration coverage for fullscreen, missing and unsupported preferences, legacy URI metadata, and result-only widgets, verifying consistent tool events and preservation across session resume.
GitOrigin-RevId: 2a9bed804dcea8c2b4903b8406fa6710681f7f5e
## Why
Packaged uninstall could leave sandbox user profiles and desktop-created data behind. Cleanup also needs to handle a service stop before package removal completes without deleting data belonging to an update or reinstall.
## What changed
- Delete sandbox profiles before their accounts, preserving accounts for retry when profile deletion fails. Defer retained runtime accounts and the sandbox group until runtime registrations are removed and profiles unload.
- Allow cleanup while the exact retiring package is still registered, while preserving desktop data when a successor package is present.
- Remove desktop-created Codex homes during registered runtime cleanup, preserve existing CLI data, and prune empty home and cache directories. Keep directory pins through retries and avoid privileged traversal after releasing the home.
- Retry cleanup up to five attempts after a service stop, while retaining shutdown cancellation behavior.
- Report cleanup outcomes in the Windows Event Log and emit final completion after registered runtime cleanup finishes. Remove the empty installation registry parent.
## Testing
Add Windows tests for profile deletion retry and account preservation, service-stop retries versus system shutdown, and PowerShell finalizer parsing and native binding compilation. Extend retained-token coverage to check account SID matching.
GitOrigin-RevId: 1aefa969e6aff9bb8e77e2c9fe9ee772230fc4fb
## What changed
- Accept `fileId` alongside the existing `url` form for app-server image inputs, and forward file references to the Responses API as `file_id`. Update generated schemas and client types.
- Preserve file references, image detail hints, and mixed inline/file image ordering through user-message events, thread history, and rollout migration. Retain file images when truncating tool output.
- Pass file references through image preparation without resolving them, while keeping resize-notice numbering correct. Omit them from unsupported TUI display and Guardian image context.
- Reject image-edit requests whose recent-image window includes a file reference, preventing selection of an older inline image instead.
## Testing
Add coverage for serialization, request and rollout preservation, mixed-image history ordering, incomplete ordering metadata, tool-output truncation, and rejection of unsupported image-edit selections.
GitOrigin-RevId: 6ca20a8577155cc934b720803c3b7b3bffdf972a
## Why
A checkpoint produced by a previous model may be incompatible with the selected Guardian reviewer. Legacy review must retain user restrictions and verified answers across compaction and restart until a compatible checkpoint is available.
## What changed
- Expose retained user evidence independently of whether Guardian reviews the legacy transcript or parent context.
- Resolve the selected reviewer's compaction compatibility hash during resume and remote compaction. Preserve the legacy transcript when checkpoint compatibility is unknown or mismatched.
- Activate parent-context review immediately after compatible compaction and invalidate pending reviews bound to the previous evidence policy. Ordinary compaction preserves pending reviews.
- Keep strict compatibility checks for previously migrated checkpoints whose complete legacy transcript is no longer available.
## Testing
Add coverage for mismatched checkpoint hashes, retained evidence during legacy review, and pending-review cancellation on migration. Add a request-history scenario covering a model switch, incompatible automatic compaction, restart, and compatible manual compaction with preserved user restrictions and verified answers.
GitOrigin-RevId: b52bfd7c21ce7ebed482174c8d7c327e50b1ada8
## Why
Older compaction checkpoints can lack a producer model hash. Enabling thread-owned Guardian context must accommodate those checkpoints without losing user instructions or verified answers.
## What changed
- Select review policy from each history snapshot independently of the session's evidence capture policy.
- Keep legacy review for checkpoints with missing or empty producer hashes when the transcript preserves the evidence. Preserve strict compatibility checks when review depends on retained-only evidence.
- Continue capturing retained instructions and answers during migration, while supplying legacy reviews with runtime answers and subagent reviews with root authorization evidence.
- Record producer model hashes on new compactions regardless of context mode.
## Testing
Add coverage for checkpoint migration across compaction and resume in both history storage modes, snapshot policy stability, retained-evidence preservation, and authorization revalidation for owning sessions and subagents.
GitOrigin-RevId: ec72b4a51d40729173e24ac1b53b7d516ddaa825
Replace `◦` with `•` in the rendered popup before comparing the
`image_preparation_disconnected` snapshot.
GitOrigin-RevId: f951c5e1f230bf8ee6d7f32a41f41316feabcb0f
## Why
`codex app-server daemon update` previously required a latest-channel installation, leaving pinned and local managed packages unable to return to production updates through that command.
## What changed
- Allow explicit updates to restore managed packages to the latest stable release while preserving the automatic-update preference. Extend legacy migration to local and pinned packages when the published installer and release support it.
- Bind restoration to the selected release with a single-use updater authorization and installer guards. Scheduled updates and ordinary updater socket requests continue to respect pins.
- Restart a running daemon when its selected package changes, even if the binary and version are identical; leave stopped daemons stopped.
- Reject incompatible production packages before selecting them, resolve Windows junction targets for ownership checks, and show the restoration command after installing a pinned CLI package.
## Testing
Extend daemon and installer tests to cover local-package migration, restoration with automatic updates enabled or disabled, same-binary package restarts, stopped-daemon preservation, selection races, incompatible releases, and socket requests that must not undo pins.
GitOrigin-RevId: d64520c1190b5c7bac084fb4a1b6b24998b08dfa
## Why
Locale, timezone, or system clock changes can alter the `ps` start-time text used to identify a running daemon, causing its PID record to be treated as stale.
## What changed
- Record and check native process identities on Linux and macOS, using boot IDs to reject records from previous boots and native process details to detect PID reuse.
- Keep `processStartTime` for older clients and promote verifiable legacy daemon and updater records when the updater starts.
- Retain legacy records and report an error when a live process's start-time text no longer matches. Propagate Windows process-access errors instead of treating inaccessible PIDs as stale.
## Testing
Add regression coverage for locale and timezone changes, altered legacy timestamps, legacy record promotion, PID reuse, previous boots, and macOS process ownership differences. Extend zombie-reaping coverage to native identities and verify that Windows access-denied errors are preserved.
GitOrigin-RevId: b5752497c0e8186bb604fe5dd33f570acc068f20
## What changed
- Expose `analytics_plan_history` in the experimental features menu for previewing consumer five-hour and weekly allowance history in `/analytics`. Keep it disabled by default.
- Add `h` and `l` as left/right navigation aliases in analytics, preserving explicit bindings and honoring remapped or unbound arrows.
- Use terminal-aware footer key colors to keep analytics shortcuts readable on light backgrounds.
## Testing
Add navigation tests for arrow equivalence, custom bindings, modifiers, and key repeat. Extend light-background contrast assertions and update style snapshots. Add a snapshot for the plan history experimental feature entry.
GitOrigin-RevId: 310e64a43a7dd08c8a3fd1efed6985805f7dd580
## Why
Reading daily totals required selecting each bar, and changing grouping required opening a picker. Section navigation could shift vertically, while scroll hints appeared even when content fit.
## What changed
- Make `g` cycle through available groupings and summary views directly.
- Show daily totals on charts spanning up to seven days when space permits, prioritizing the selected value and keeping neighboring labels separate.
- Keep section navigation at a stable height, show more legend rows in taller maximized reports, and move their update timestamp into the footer.
- Hide scroll hints when content fits, including an exact fit, and show only the active summary view label.
## Testing
Add regression tests and snapshots for seven-day totals, narrow label spacing, grouping wraparound and server capabilities, configured `g` navigation in ungrouped sections, stable tab placement, adaptive legends, footer timestamps, and overflow-dependent scroll hints.
GitOrigin-RevId: 9d14f314b4a22ca2d926f0e69c6b89e4f9c8550d
## What changed
- Make Summary the initial Analytics tab, showing profile identity, token totals, streaks, activity insights, and most-used plugins and skills.
- Move daily, weekly, and cumulative token activity charts into Summary. Replace the usage menu's separate views with “View analytics”; `/usage weekly` and the other explicit modes open Summary with the requested view, while reopening from the menu retains navigation.
- Load account profiles independently of other reports, preserve missing values separately from zero, and reject responses when the active account changes. Refresh retries profile failures.
- Adapt Summary to narrow terminals and compact large chart totals and axis labels. Show the account email without its ID in the Analytics header.
- Flush queued transcript history before opening an overlay so startup output stays in the transcript.
## Testing
Add profile decoding and account identity tests, Summary layout and navigation snapshots, refresh and cancellation coverage, and regression tests for retained navigation and transcript history when opening Analytics.
GitOrigin-RevId: 6dbe6c4c98209000d8b4bc18a4e01b98fab5991a
## Why
Consumer analytics lacked a per-chat view of allowance usage and balance credit debits.
## What changed
- Add a Top chats panel for up to 100 local chats active in the past 30 days, including discovered descendants and excluding archived roots.
- Show weekly and five-hour usage as percentages of current full limits, plus balance credits including adjustments. Preserve exact credit decimals and distinguish missing values from zero.
- Support cycling sort metrics with `s`, a top-five dashboard summary, and expandable model, reasoning effort, and speed breakdowns.
- Query task usage through `query_v2` with bounded, disjoint descendant groups. Retain partial and unavailable rows, hide titles when usage is unavailable, and report ranking coverage and freshness.
## Testing
Add backend and TUI tests covering decimal precision, request and response validation, paginated archived descendants, unavailable responses, metric selection, title hiding, and top-five rendering.
GitOrigin-RevId: b8f233166576389c8212bb97eceb7c6be7f5f51d
## What changed
- Add consumer plan usage history behind `analytics_plan_history`, disabled by default, fetching seven days of five-hour and weekly allowance periods.
- Show period usage percentages with expandable breakdowns by feature, model, surface, or turn start, plus snapshot freshness and incomplete coverage indicators.
- Preserve unknown usage separately from zero and handle an unavailable history endpoint without blocking other analytics reports.
## Testing
Add backend and TUI tests for unavailable endpoints, response validation, feature and account gating, account changes during requests, period navigation, snapshot resets, and narrow and wide layouts.
GitOrigin-RevId: 7bea87a3407730b18040d7dda7e7dd0afe67fb2e
## What changed
- Show local chats active in the past 30 days, ranked by estimated lifetime credits, for supported Business and Enterprise plans. Exclude archived chats and subagents.
- Add expandable model, reasoning effort, and speed breakdowns, optional zero-credit groups, and dollar estimates when supplied by the backend.
- Load estimates in bounded batches of up to 100 distinct threads, retaining available results when some estimates are unavailable. Hide chat titles without a returned estimate and distinguish missing usage from zero.
## Testing
Add coverage for batch validation, partial failures, request limits, pagination and ranking, plan eligibility, title visibility, and responsive table navigation and details.
GitOrigin-RevId: 783b960a780f4e0fd39f57e2fa78fb31b3090c44
## What changed
- Add “Explore analytics” to the `/usage` menu for ChatGPT accounts, opening a full-screen dashboard.
- Show usage and messages for consumer accounts, credits and token usage for Business and Enterprise accounts, and plugin and skill activity for both.
- Support overview and maximized views, 7- and 30-day ranges, report grouping, daily details, model filtering for token usage, and refresh.
- Keep report loading and errors independent, cancel pending loads on close, and retain navigation state across reopening while preserving the composer draft.
## Testing
Add regression and snapshot coverage for account-specific reports, range and grouping changes, model filtering, loading and cancellation, small terminals, terminal colors, and draft preservation.
GitOrigin-RevId: 46da94a14d6e227551e740b85f9d2d552aaf75b0
## What changed
Add daily stacked bar rendering in the TUI analytics module, staged for dashboard integration.
- Plot three categories plus an Other remainder without renormalizing values, with separate positive and negative bands for signed credits.
- Add readable numeric axes, calendar ticks, a selection cursor, and a callout showing the authoritative daily total. Distinguish missing days from zero usage.
- Keep the selection visible in narrow viewports and omit labels that cannot fit without truncation.
- Use terminal palette colors for series and adapt secondary text to terminal color support and background.
## Testing
Add unit and snapshot coverage for axis scaling, signed bars on light and dark terminals, missing and zero values, narrow layouts, tiny segments, authoritative totals, and duplicate-date remainders.
GitOrigin-RevId: 2ed58def17abd14e77929016a54fd20e91d45670
## Why
Token plan claims can be stale, and the active account or user can change while a request is in flight. Analytics report selection and account-bound response data need to reflect the verified identity and current server plan.
## What changed
- Fetch the active account's plan once per analytics session and use it to select report endpoints and supported credit breakdowns.
- Add report loading with a fixed end date, account-scoped response caching, and token model filtering. Reuse payloads across grouping changes and evict invalid cached responses so requests can retry.
- Prefer complete attribution for usage breakdowns within the requested range; retain legacy surface/model data when attribution is incomplete and include all features in turn-start breakdowns.
- Add cancellable report-loading state with timeout and interruption errors, and preserve actionable sign-in and retry messages.
- Recheck the active identity after rate-limit reads before exposing account-bound fields.
## Testing
Add regression tests for server plan discovery, report routing and caching, model filtering, attribution fallback, failed-request retries, load cancellation, authentication recovery, and account or user changes during requests.
GitOrigin-RevId: 45c4c09f108c9703893f3ed15613437ebd7c74a8
## Why
The pinned V8 can inline `Array.prototype.sort` with incompatible element kinds when a comparator mutates the array, allowing an object to be stored in an integer-elements array.
## What changed
Disable Maglev, Turbolev, and TurboFan array builtin inlining during code-mode runtime initialization until the V8 artifacts include the upstream fix.
## Testing
Add an integration regression test that requests top-tier and Maglev optimization, checks element kinds after comparator mutation, and verifies ordinary numeric sorting still returns `[1,2,3]`.
GitOrigin-RevId: cea38e920245d6a133a5263118dc664fb3e61838
## What changed
- Track the Windows sandbox implementation separately from its legacy setup level.
- Select dedicated proxy listeners for `WindowsMxc` on Windows across sessions, app-server command execution, and the sandbox CLI.
- Rebuild the session proxy when routing changes, retaining its network policy decider.
- Add a sandbox CLI execution path through the MXC sandbox transform when `WindowsMxc` is selected.
## Testing
Extend configuration assertions for the selected sandbox type and the session refresh test to verify dedicated routing and policy decider retention on Windows.
GitOrigin-RevId: 2a22192636c542022ec81a378a16c1bb6867e145
## Why
Disabling a connector's canonical plugin could leave its tools available when another enabled plugin contributed the same connector.
## What changed
Retain `canonical_app_id` in the remote installed-plugin cache and use ownership metadata for the current account when building connector snapshots. A disabled canonical owner now excludes its connector even when another plugin contributes it or the owner's bundle is absent from the host. Match owners by plugin name and marketplace, and ignore ownership metadata when the cache no longer matches the current authentication.
Build these snapshots through `PluginsManager` when the plugins feature is enabled; otherwise use an empty snapshot.
## Testing
Add manager coverage for combined local and canonical exclusions, marketplace matching, and authentication changes. Add an integration test showing that disabling a noncanonical contributor preserves shared calendar tools, disabling the canonical owner hides them, and clearing the exclusion restores them.
GitOrigin-RevId: ee2981d1277825fefb671dd7d1078cd7ea5c0ea5
Expose `shell_startup_script` from `codex-shell-command` and use it for
interactive shell snapshot capture, preserving the existing startup scripts.
The helper returns an empty script for other shell types; callers remain
responsible for login startup flags and shell lifecycle.
GitOrigin-RevId: 8cb64c02c08140d2676e67b9db06c9a77649b3eb
## What changed
- Add `AnalyticsSession` to load local ChatGPT credentials, bind requests to the initial account and user, and reject identity changes before requests or when accepting results.
- Reload credentials and provide bounded recovery for unauthorized requests, using a backend client that disables redirects.
- Add lazy TUI analytics session initialization, account display metadata, and plan-specific credit groupings for future dashboard integration.
## Testing
Add regression tests for account and user changes, missing or API-key-only authentication, credential reloads, recovery after a `401` response, and session initialization after signing in.
GitOrigin-RevId: 2ba0d96fb44936d12d13aedffc3f2e61aedea560
## What changed
- Add daily text token history grouped by model or token type, with date-range and model filtering, freshness metadata, and validation of token counts. Support both grouped and per-model report data, including reported model totals when components are unavailable.
- Add credit formatting helpers that preserve tiny signed adjustments and format integer millionths without floating-point rounding, plus a short date formatter.
## Testing
Add unit tests for token groupings, model filtering, freshness, missing components, explicit zero totals, and credit formatting across tiny refunds and integer limits.
GitOrigin-RevId: 706dcae7995e46707c2f1df9cdf3e06966459dd6
## What changed
Add display models and adapters from typed backend responses for usage, credits, messages, plugins, and skills. Normalize reports into daily histories with grouping labels, units, and freshness timestamps.
Preserve missing values, signed credit adjustments, and full daily totals when grouping usage by task start. Account for unassigned message counts as `Other`, merge duplicate dates, and fill unreported days only for credit reports. Group models contributing less than 1% of relative usage into `Other`.
This adds the data preparation layer for dashboard integration; it is not yet connected to `/usage`.
## Testing
Add unit tests covering missing versus explicit zero values, signed credits, enterprise labels and freshness, usage grouping, duplicate-date message remainders, and unsupported message groupings.
GitOrigin-RevId: 931f46a9a6521b4cda2fc26051e7a305cbf6fa6c
## What changed
Add `Client::get_account_analytics` with public `AnalyticsReport`, `AnalyticsResponse`, and analytics models for token usage, credits, workspace messages, plugins, and skills. Select report-specific routes and query parameters for both `api/codex` and `wham` paths.
Preserve backend attribution, optional amounts, and signed credit values. Keep string values open for forward compatibility, and return a generic decoding error without including the response body.
## Testing
Add HTTP contract tests covering report routing and decoding, plugin query parameters, signed credit events, new string values, and rejection of mismatched response shapes without exposing the body in the error.
GitOrigin-RevId: 196ff7f49c3f9d8a029398d09af14449c5d04207
## Why
`allowedWindowsSandboxImplementations` uses the legacy setup-mode type, which cannot represent `mxc`.
## What changed
- Introduce `WindowsSandboxImplementation` with `elevated`, `unelevated`, and `mxc` variants for configuration requirements, and update the generated schemas and TypeScript and Python types.
- Keep `WindowsSandboxSetupMode` limited to `elevated` and `unelevated`, and update the requirements mapping and its test assertions to use the new type.
- Document that clients selecting `mxc` skip the legacy setup and readiness APIs and use the standard `command/exec` streaming and process-control path.
GitOrigin-RevId: c926e853f0bbdb29d4b437866c284c1394a7ea13
## What changed
- Replace dedicated Guardian endpoints with `/responses`, sending `x-codex-guardian: reviewer` or `x-codex-guardian: classifier` for eligible Codex backend requests over HTTP and WebSocket.
- Add model-scoped thread headers, recheck backend authentication on each request attempt, and reconnect WebSockets when the applicable headers change.
- Retain `features.guardianv2.free_guardian` for configuration compatibility while removing its routing gate; the backend now controls Guardian billing.
## Testing
Update tests for reviewer and classifier headers, model and authentication scoping, HTTP fallback, WebSocket reuse, and parent-response metadata across retries.
GitOrigin-RevId: 1c2b3c458ab77fb40aae4ee6784f6d827c60e76c
## Why
MXC is a sandbox implementation, not a restricted-token sandbox level. Executor requests need to represent that choice separately from `WindowsSandboxLevel`.
## What changed
- Introduce `WindowsSandboxSelection` for executor sandbox contexts and remove `Mxc` from `WindowsSandboxLevel`.
- Preserve the `windowsSandboxLevel` wire field and its serialized values for compatibility.
- Share sandbox selection between executor process launches and filesystem helpers, and use the new selection in capability discovery and skill reads.
- Disable Windows sandbox selection for executor paths that do not use Windows path conventions.
## Testing
Extend coverage for MXC wire serialization, Windows skill-read sandbox checks, and capability discovery with distinct permissions. Exercise remote filesystem write restrictions with both restricted-token and MXC sandboxes, including rejection when native MXC is unavailable.
GitOrigin-RevId: 266211377bcb138a0dc75861e9ff2225fa37a53d
## What changed
Move concrete reviewer settings into `guardian-v2` and replace `ReviewerConfigOverrides` with a `ReviewerConfig` callback stored in thread extension data. Apply the callback to each captured parent configuration before preparing context and checking session reuse.
Keep model selection, policy prompt construction, and live network state in the core adapter. Preserve the existing read-only reviewer settings and share `reviewer_permission_profile` with inherited environment configuration.
## Testing
Update existing Guardian configuration, session reuse, and prewarming tests to use the extension's actual configuration builder, including compiling the shared source in core's test host.
GitOrigin-RevId: dfa1dae9f8d4469f2e4f6e64ab842cc662970898
## What changed
- Record `auth_or_link` for Codex app calls with trusted connector authentication failure metadata, capturing it before result callbacks or model-facing rewrites. Propagate the classification to MCP tool-call and app-usage analytics.
- Record `approval` when a dispatched approval request is denied, times out, or aborts, covering both MCP elicitation and legacy user-input approvals.
## Testing
Add unit and turn-level coverage for authentication failures, metadata removal by callbacks, successful and rejected approvals, and closed approval response channels. Verify event counts, classification, and exclusion of sensitive test values and authentication URLs from analytics payloads.
GitOrigin-RevId: 493df2a2923f346a0f644f1f26ab8b6b9b8888e0
## What changed
- Add `AllowedTools` to the extension API, captured once at thread startup to restrict which tools can be advertised or executed. An empty list disables all tools; an absent value preserves ordinary tool setup. Callers must supply the list again when resuming a thread.
- Filter tool registration and hosted tool specifications before Code Mode and discovery. Match names with their namespaces and require generated tools such as `exec` and `wait` to be explicitly allowed.
- Move Guardian reviewer tool selection into the allowlist while retaining feature and sandbox restrictions, with a fallback for older reviewer sessions.
## Testing
Add coverage for selected, empty, and absent allowlists across tool sources and Code Mode. Add managed-thread tests verifying advertised tools, rejection of excluded calls, and that replacing extension state after startup cannot change the captured allowlist.
GitOrigin-RevId: 33c1722c4241f368963d8385fa0415112e16a72d
## What changed
Introduce `ReviewRequest` in `codex-guardian-reviewer` to own contributor routing, cached approvals, synchronous fallback, and review cancellation. Keep action validation and session-specific preparation in the host.
Track the complete approval operation through parent shutdown, driving cancellation through reporting and reviewer cleanup before releasing it. Preserve fresh-review requirements and validate cached approvals before recording their outcome.
## Testing
Extend coverage for cancellation before routing and during cached approval, parent shutdown cleanup, and required fresh review overriding a cached allow result with matching assessment events and a denial warning.
GitOrigin-RevId: 8dcf26cc3660e91ef94e28e0bee7327ebe34ae6c
## What changed
Replace duplicated model-selection fields in `GuardianReviewSessionConfig` and `GuardianReviewSessionParams` with the existing `ReviewModel` struct. Read model settings and selection metadata from that struct for session execution, analytics, and failed-review records, and update the existing test fixtures accordingly.
GitOrigin-RevId: 1da2001e68168ec40a3d726694eb5fd0445aa3d7
## What changed
Remove core guardian tests for transcript numbering and unpaired tool outputs. Extend `guardian-context` coverage to assert that named image-only tool outputs retain their source and use `[non-text output]` for both sync and async contexts, with tool calls included or excluded.
GitOrigin-RevId: ba483e8f6e335697106a987051687a12ad2fde01
## What changed
Make `parse_guardian_assessment`, the rejection circuit breaker types and methods, and `AUTO_REVIEW_DENIAL_WINDOW_SIZE` crate-visible with `pub(crate)`. Update their re-exports in `guardian-reviewer` to match.
GitOrigin-RevId: 39521c7859936ebc9af570d7f1ad7b2348f3673d
## What changed
- Mark `guardian_ext` as a removed compatibility flag.
- Remove `InternalSessionSpawner`, `InternalSessionSpawnFuture`, `ApprovalReviewInput`, and `ApprovalReviewError` from the extension API, along with the spawner test.
- Remove `NodeReplReviewEvidence::review_inputs` and make the core re-exports of `NodeReplReviewEvidenceMode` crate-private.
GitOrigin-RevId: 1c532bc57264644cf521ef8077285a9b8ed67391
## What changed
- Make `SynchronousReview` own assessment events, telemetry, warnings, evidence-recording decisions, and consecutive-denial accounting.
- Keep action preparation, stale-approval validation, evidence storage, and event publication in the host adapter.
- Extract `Session::interrupt_turn_with_warning` to apply extension-requested interruptions to the selected active turn and emit the interrupted thread-idle lifecycle event.
## Testing
Adapt session tests to verify that extension interruptions emit the thread-idle lifecycle event and survive shutdown of the calling runtime.
GitOrigin-RevId: b234ad86206fedc2434f48fc44dc57679443c69f
## What changed
Extract V2 interruption validation and dispatch into `AgentControl::interrupt_spawned_agent`, returning the agent path and previous status with typed validation and runtime errors. Update the `interrupt_agent` tool handler to delegate to it while retaining tool error mapping and activity emission.
Preserve rejection of root and self targets and successful handling of unloaded or already-dead runtimes without reloading them.
GitOrigin-RevId: 893bbc4f464596d0fee032a71702a4a4f6becf5c
## What changed
- Use cancellation guards to tie reusable reviewers and temporary forks to their lifetimes, including when a review future is dropped. Replace cancelled reusable reviewers before reuse.
- Share `ReviewerTasks` between the pool and `ThreadManager`, and wait for tracked cleanup during pool shutdown. Remove the separate session shutdown protocol.
- Move denial cleanup from core task handling into Guardian's turn start, stop, and abort hooks. Expose `install_reviewer` to register both thread and turn lifecycle contributors together.
## Testing
Add an integration test that exercises stale denial cleanup at turn start, completion, and interruption, including starting another turn after interruption.
GitOrigin-RevId: 85badb235309c1e1576547d1746f7ee9fc66d545
## What changed
Extract target validation, runtime reloading, and message delivery from the V2 tool handler into `AgentControl::deliver_message`. Keep target resolution, analytics, and tool-facing error mapping in the handler.
Represent plaintext and encrypted payloads with `AgentMessage`, sharing communication construction with agent spawning. Preserve queue-only and follow-up turn semantics, target checks before reload, and turn metadata propagation.
GitOrigin-RevId: a88bcede015a17f6c11c7cad95ee58add5dc2820
## What changed
Move shared spawn and resume configuration helpers into `agent::child_config` and route both multi-agent versions through `prepare_agent_spawn_config`.
Preserve each version's role and model precedence, full-history inheritance, service-tier selection, runtime policy, and default-role metadata for reloads. Return configuration errors as strings and convert them to `FunctionCallError::RespondToModel` at tool-handler boundaries.
GitOrigin-RevId: 9d539050aa44f94043a6487f1f6f1037de1d15ca
