## What changed
- Scan up to eight skill roots concurrently instead of loading them one at a time.
- Share scan capacity across host-skill and plugin-skill loads to bound aggregate filesystem work.
- Merge completed scans in input order so root precedence and error ordering remain deterministic.
- Skip canonicalization and parsing work for empty roots.
## Testing
- Add coverage for out-of-order scan completion and shared-capacity blocking.
GitOrigin-RevId: 33a47aaa02aee7fe31a050bb04d51c1e2043d6ae
## Why
Queue-only child-agent updates that arrive around a final answer should not
restart sampling, but they must remain available to the next turn. Explicitly
injected response items still need to reopen the current turn.
## What changed
- Defer queue-only inter-agent mail after a final answer without treating it as
same-turn pending input.
- Re-enable current-turn mailbox delivery when the model requests a follow-up,
a stop hook continues the turn, or a response item is explicitly injected.
- Check for turn-triggering mailbox work after clearing the completed active
turn so pending work can start safely.
## Testing
Added coverage that queued child mail waits for the next user turn and that an
injected response item after a final answer triggers a follow-up request.
GitOrigin-RevId: aa81d00707943284b7ab069b60970e547908c895
## What changed
- Allow `thread/start` to select experimental `historyMode: "paginated"` when the thread store supports turn and item pagination.
- Resume paginated threads with bounded model context and require `excludeTurns: true` instead of loading full history.
- Return `turnsBackwardsCursor` and `itemsBackwardsCursor` from `thread/resume` so clients can hydrate durable history through `thread/turns/list` and `thread/items/list` while newer records arrive through live notifications.
- Reject operations that still require full history, including `initialTurnsPage`, `thread/read` with `includeTurns: true`, `thread/rollback`, and detached review.
- Make item cursors independent of the optional `turnId` filter.
## Testing
- Cover paginated thread start, cold and live resume, backwards cursor hydration, metadata reads, unsupported operations, and stores without pagination support.
GitOrigin-RevId: c42f9b268e29d049843330f474fa3ecde12f5b8a
## Why
Cached MCP tool catalogs must not substitute stale session state for the live
server connection.
## What changed
- Verify `mcpServerStatus/list` waits for replacement stdio server metadata
before combining it with cached tools.
- Verify configurations with remotely sourced environment variables bypass the
tool catalog cache.
- Assert that calls made from cached definitions return output from the live
replacement process.
- Make the stdio test server available to Windows Bazel integration tests.
GitOrigin-RevId: 0e518cd46f0d89af28e3a4e256023b5872e6e2fb
## What changed
- Recognize the experimental `codex/tool-catalog-cache` server capability during MCP initialization.
- When the server advertises `cacheable: false`, clear its shared tool catalog snapshot and prevent subsequent fetches from repopulating the cache.
GitOrigin-RevId: bce98371a68881001d3ab1378fa1548575f2dadc
## What changed
- Add realtime conversation version `v3`, which preserves the V1 Codex Voice behavior while using Frameless Bidi `delegation.*` events.
- Translate audio, transcripts, handoffs, session context, and lifecycle events between the app server and the Frameless Bidi wire protocol.
- Support `v3` over WebSocket and WebRTC, including the Frameless `/live` endpoint, session configuration, headers, and default model selection.
- Update the app-server protocol schemas and documentation for the new version.
## Testing
- Add unit coverage for Frameless event parsing, outbound messages, context chunking, URL construction, and call creation.
- Add app-server end-to-end coverage for WebSocket delegation and WebRTC session startup.
GitOrigin-RevId: 79a3307bc209e1a54582ebd2febc07c909fca016
## Why
Token-budget sessions need a final opportunity to preserve important state before the current context is compacted.
## What changed
- When `auto_compact_fallback_prompt` is configured, reserve `auto_compact_fallback_buffer_tokens` beyond the base auto-compaction limit.
- At the base limit, inject the fallback prompt once as a developer message while keeping the normal tool surface available. Continue reporting zero base-window tokens during this reserve.
- Roll over when the reserve or full model context is exhausted, and skip the fallback when a new context was already requested.
## Testing
Added coverage for using the fallback reserve, rolling over after it is exhausted, and bypassing the fallback for an explicit `new_context` request.
GitOrigin-RevId: 32983ecadc5cb4927e15ecd7b1e896cb52a9ad3b
## What changed
- Track connected and disconnected states across initial remote exec-server connections and reconnection attempts.
- Emit experimental `thread/environment/connected` and `thread/environment/disconnected` app-server notifications for each thread selecting the environment. Each payload identifies the thread and environment; current state is not replayed when a thread starts.
- Stop forwarding connection events when an environment selection is removed or replaced.
## Testing
- Cover connection, disconnection, reconnection, shared-environment notifications, and replacement of a selected environment.
GitOrigin-RevId: 5dd767372363c4a2a8319fc16164be117d5bd20c
## What changed
- Add `auto_compact_fallback_prompt` and
`auto_compact_fallback_buffer_tokens` to `features.token_budget` and the
generated configuration schema.
- Trim empty fallback prompts, limit prompts to 2,000 bytes, require a buffer
when a prompt is configured, and reject non-positive buffer values.
- Preserve the new settings when locking resolved session configuration.
## Testing
- Cover config resolution and validation for overlong prompts, missing
buffers, and non-positive buffers.
- Extend the configuration-lock test to cover both settings.
GitOrigin-RevId: 6463f963ba1dbd633d76601b8344360a8c85cf8b
## Why
Tabs in composer text could render at a different width from the cursor and
wrapping calculations, causing misaligned cursors and unexpected line breaks.
## What changed
- Replace tabs with single-column spaces for wrapping and rendering while
preserving the original editable text and byte ranges.
- Apply the same display representation to base text, styled elements, and
highlighted ranges.
## Testing
Add coverage for cursor alignment, rendered output, and wrapping of text that
contains tabs.
GitOrigin-RevId: 06d8489621d2f8b0823c7a004079d152dd0cab66
## What changed
- Use `$`-mention boundary rules when deciding whether a skill mention followed by a suffix is already bound.
- Fall back to the editable skill token on the left when the token on the right is a bound mention such as `$bound/path` or `$bound.config`.
## Testing
- Add completion-target coverage for path and configuration suffixes.
- Add a composer snapshot covering the skill popup fallback for a bound path suffix.
GitOrigin-RevId: bf72237dfd56cfe6f590a3fab10c2bfe03b088a8
## Why
Skill mentions and shell parameters both use a `$` prefix. Numeric and special
shell parameters should not open or displace completion popups, while loaded
skills can legitimately have digit- or hyphen-leading names.
## What changed
- Classify `$` queries as skill candidates, shell variables, definite shell
parameters, or ambiguous shell-like names.
- Offer completion for an ambiguous name only when it fuzzy-matches a bindable
loaded mention, reusing the filtered catalog when the popup opens.
- Keep definite shell forms such as `$1`, `$12`, `$-`, and `$_` out of mention
completion and preserve completion targets elsewhere in the composer.
## Testing
Added unit and snapshot coverage for positional and special parameters,
digit- and hyphen-leading skills, uppercase shell variables, unbindable
qualified skills, and competing completion targets.
GitOrigin-RevId: 8bb64cb5ca0fa8233c7fadb11a53678d57eb814c
## What changed
Add named tracing spans around step environment snapshots, capability and
executor plugin resolution, `AGENTS.md` refreshes, and MCP runtime projection
and refresh operations. Skip recording function arguments to keep these spans
focused on timing and execution flow.
GitOrigin-RevId: 73a452ef1b5da6f7f2b00d24421a815c5d514eaf
## What changed
- Add experimental `beforeTurnId` support to `thread/fork` for copying history strictly before a turn, including an in-progress turn.
- Use before-turn forks for TUI backtracking and safety retries so the replacement prompt retains thread context instead of recreating the branch from the preceding turn or a new thread.
- Add experimental `deferGoalContinuation` support so safety-retry forks can inherit an active goal while waiting for the replacement turn before automatic continuation resumes.
- Hide duplicated nested-review prompts when replaying history or selecting a prompt to edit.
## Testing
- Cover before-turn truncation, fork validation, active-goal inheritance, TUI safety retries, backtracking, and nested-review replay.
GitOrigin-RevId: 4915d85ae1929a0b8c40f3ed7dca29c7a5bdcdf3
## What changed
- Replace the combined `Codex` wrapper with an `Arc<Session>` and a dedicated
`SessionIo` handle for submissions, events, status, and loop termination.
- Make `Session::spawn` return the session and I/O handle separately, and have
`CodexThread` own both.
- Update thread management, delegated sessions, guardian reviews, and agent
control to use the appropriate session or I/O interface.
GitOrigin-RevId: 8523bad7e7366e2969fd0cc3849b4247b141427a
## Why
Retrying a safety-buffered turn by rolling back the active thread rewrites the
source conversation and requires the TUI to repair its local transcript state.
## What changed
- Interrupt the buffered turn, then fork after the preceding completed turn and
submit the prompt on the new thread with the faster model and low reasoning
effort.
- Start a new thread when the buffered request is the conversation's first turn.
- Preserve committed steering input in the retry, leave the source thread
intact, and restore the prompt alongside any existing draft if branching
fails.
- Do not offer the retry action for side conversations or stale and in-progress
turns.
## Testing
Added coverage for retries with prior context, first-turn retries, committed
steering input, branch failures, existing composer drafts and attachments, and
side conversations.
GitOrigin-RevId: adb49da8236e9212ef0082e033e6f715986b5cd9
## What changed
- Add an explicit restore mode for thread input snapshots and use it when replaying thread snapshots.
- Preserve the pending-steer interrupt flag alongside running-turn and queued-input state when restoring an in-flight turn.
- When in-flight state is not preserved, clear stale running flags and move pending steers ahead of already queued follow-ups while retaining their history records.
## Testing
- Expand TUI tests to cover both restoration modes, including sleep-inhibitor state, pending-steer ordering, history records, and queued submission.
GitOrigin-RevId: f4e8b2bdf065ef6c6644b7f0ea9cf32c8a9af619
## What changed
- Fork the conversation immediately before the selected prompt instead of rolling back the current thread in place, preserving the original conversation.
- Reopen the selected prompt in the new conversation with its images, text elements, and mention bindings intact.
- Start a fresh conversation when editing the first prompt, show a dedicated continuation notice, and restore the prompt with an error if branching fails.
## Testing
- Cover branching before first and later prompts, source preservation, prompt restoration, hidden review prompts, mid-turn steering, in-progress turns, stale transcript entries, and mention bindings.
GitOrigin-RevId: 15473535b3f9733b0300711a498969e6d389db79
## Why
Core filesystem permission models operate on native absolute paths, while exec
sandbox contexts must serialize paths as portable URIs.
## What changed
- Make the core filesystem permission and profile types concrete over
`AbsolutePathBuf` instead of generic over their path representation.
- Add exec-specific permission types that convert native permission paths to
`PathUri` values at the sandbox protocol boundary.
- Replace the legacy read/write-roots tuple with a named
`LegacyReadWriteRoots` struct.
## Testing
- Add coverage that exec sandbox permission paths serialize as URIs and round
trip through the filesystem protocol.
GitOrigin-RevId: a0ef5f8aba9bbfdf00a00fcc4f199314836bc85d
## What changed
- Leave an output-free interrupted prompt in the transcript and open a blank
composer for the user's next instruction.
- Use the normal interruption notice for both `Esc` and `Ctrl-C` interrupts.
- Track safety-buffering prompts separately so failed retries can still restore
the prompt, including after switching between threads.
## Testing
- Cover output-free `Esc` and `Ctrl-C` interrupts, failed safety-buffer retries,
and restoration of safety-buffer state from thread snapshots.
GitOrigin-RevId: 30b368d00dfdb981d06d4c95ce3f252389b7917c
## Why
Sparse and out-of-order rate-limit updates can omit account metadata or let an
older full read overwrite a newer workspace hard stop. Credit availability also
does not always include a displayable balance.
## What changed
- Propagate the backend's `spendControlReached` state through rate-limit
snapshots and the app-server protocol, preserving it across sparse updates.
- Keep workspace hard-stop classifications for error routing and ignore stale
full-read results after a newer hard-stop notification.
- Base TUI warnings and model-switch prompts on credit availability and hard-stop
state, and show `Credits: Available` when a finite balance is hidden.
## Testing
Added coverage for spend-control mapping, snapshot merging, out-of-order reads,
workspace error routing, warning and prompt behavior, and credit status display.
GitOrigin-RevId: 18b09d493d5a620a7c079fb721ec50b38dd5b04c
## What changed
Resolve approval test targets outside the workspace relative to the test's
temporary home instead of the process working directory. This keeps the targets
outside the test workspace while containing them in the test environment.
GitOrigin-RevId: 871e840d658468c56d603afc10cc226f918bba5a
## Why
Starting a new session should not have to wait for an unchanged stdio MCP
server to initialize before its tools can be presented to the model.
## What changed
- Cache recent tool catalogs for matching stdio MCP server configurations and
use them while a new live connection starts in the background.
- Bound the process-wide cache to 32 entries with a 30-minute TTL, and key
entries by the server configuration, environment, working directory, and
elicitation capabilities that affect the catalog.
- Keep connection-specific instructions and tool annotations out of cached
catalogs. Resolve tool calls against the live connection so tools that are
unavailable or hidden in the new session cannot be invoked from stale data.
- Publish only the newest completed catalog fetch and do not cache HTTP servers
or stdio configurations that depend on remotely sourced environment values.
## Testing
- Add coverage for cache expiration, environment changes, catalog sanitization,
and out-of-order refreshes.
- Add an end-to-end test that starts inference from cached definitions while a
replacement MCP process initializes, then verifies calls and visibility
against the live process.
GitOrigin-RevId: 2ed0f29ba53ace12532730f951b175c588aa3ae7
## What changed
- Add `subErrorType` to external agent config import failures and carry it through progress and completion notifications, persisted import histories, and analytics.
- Expose plugin store failure subtypes from `PluginInstallError` so manual installs and imported plugins report details such as `failed_to_copy_plugin_file` while failures without a subtype remain `null`.
- Keep existing persisted import records compatible by defaulting the new field when it is absent.
## Testing
- Cover store I/O subtype extraction, analytics serialization, persisted import histories, and migration notification fixtures.
GitOrigin-RevId: ec1a85623fc826cbd7cfe9f55c2dd2bc86665937
## What changed
- Guard executor-backed MCP stdio sends with a single-permit semaphore so a
second JSON-RPC message cannot start writing while the first write is pending.
- Add a regression test that drives two concurrent sends and verifies the
process receives each newline-delimited message in order without overlap.
GitOrigin-RevId: 30c2e51827ecf57b5eb23d9ad1e876734d6e784e
## What changed
- Add an optional `policy_template` field to auto-review model messages.
- Build Guardian instructions from the catalog template when present, while
preserving the existing precedence for managed and catalog policy content.
- Fall back independently to the bundled template and bundled policy, and
preserve explicit empty catalog values.
## Testing
- Cover missing and empty template deserialization, policy precedence, bundled
fallbacks, and Guardian prewarm instructions.
GitOrigin-RevId: 93116dadbc9bf40527882844a428caed4159fff6
## Why
Amazon Bedrock can use either a Codex-managed API key or credentials managed by
AWS, so logout must not remove or misrepresent credentials that Codex does not
control.
## What changed
- Remove the managed Bedrock key on logout and clear `model_provider` only when
its user-config value is still `amazon-bedrock`.
- Reject logout for AWS-managed Bedrock credentials without changing existing
authentication or configuration.
- Add a `test-logout` app-server test-client command that waits for the
resulting `account/updated` notification.
## Testing
Add coverage for managed and AWS-managed Bedrock logout, concurrent provider
changes, config reload failures, and conditional user-config cleanup.
GitOrigin-RevId: a003c0ec27ad7b36d499fe7b63ab4a0dd6369b4f
## What changed
- Hide `gpt-5.4` and `gpt-5.4-mini` from model selection and direct existing users to `gpt-5.6-terra` and `gpt-5.6-luna`, respectively.
- Use Terra for memory consolidation and Luna for memory extraction and rate-limit switch prompts.
- Keep Amazon Bedrock GPT-5 catalog entries visible when their bundled model metadata is hidden.
## Testing
- Add coverage that all Amazon Bedrock GPT-5 models remain visible and update TUI expectations for the new model choices.
GitOrigin-RevId: fd69c8d5fb6b510c77204df6b7ac3f07e67cded2
## What changed
- Handle experimental `account/login/start` requests with `type: "amazonBedrock"`.
- Validate the API key and Mantle region, persist the managed credential, select the `amazon-bedrock` model provider, and emit the standard login and account-update notifications.
- Reload configuration for account and authentication status reads so the newly selected provider is visible immediately, with the startup configuration as a fallback.
- Extend the app-server test client with an Amazon Bedrock login mode that redacts API keys from request logs.
## Testing
- Add app-server coverage for successful login, credential and provider persistence, notifications, refreshed account state, invalid inputs, conflicting configuration, forced ChatGPT login, external authentication, and configuration reload failure.
GitOrigin-RevId: bf3ae98375caee41e78d2cdc677127138184c17a
## What changed
Document that `exec_command.yield_time_ms` has an effective range of
2,000–30,000 ms on Windows, and update the tool specification test to match.
GitOrigin-RevId: aa20bbb2ad82d162d29d93f3bfbe1b970445f44e
## What changed
- Replace pending WebSocket URL registration with deferred Noise environment
registration that gates connection attempts on an explicit readiness signal.
- Reuse the Noise rendezvous transport after readiness so reconnects request a
fresh connection bundle.
- Preserve terminal errors for failed or dropped registrations and keep late
completion isolated from replacement environments.
## Testing
- Add coverage for readiness gating, registration failure and replacement,
eager Noise connections, and reconnection through a fresh rendezvous bundle.
GitOrigin-RevId: 83e23fa03a02e3b2bdf1a83fe26d7ac461f55cf3
## What changed
- Define `SleepItem` in `codex-extension-items` and carry `clock.sleep` events through `TurnItem::Extension`.
- Preserve sleep persistence, thread-history reconstruction, and the app-server's `type: "sleep"` schema while reusing the extension-owned type.
## Testing
- Add serialization coverage for the `clock.sleep` extension wire shape and its protocol wrapper.
- Update sleep lifecycle and thread-history tests to exercise the extension-owned item.
GitOrigin-RevId: 34bb33ed9b90a09bf4744aafdb56e019c717a168
## Why
Detached reviews should behave like ordinary forked turns so clients get the normal steering, tool, permission, and item-stream behavior.
## What changed
- Add a bundled `$review-agent` skill with read-only, defect-first review guidance.
- Start detached reviews through `AgentRunner` with a target-specific prompt that explicitly references the bundled skill. Keep inline reviews on the existing review-mode flow.
- Continue emitting the detached review thread and turn identifiers through the app-server API.
## Testing
- Update the detached review integration test to verify the forked turn prompt, ordinary turn completion, and selection of the bundled skill when a user skill has the same name.
GitOrigin-RevId: 979664cd80800fb0d51eabdc4e3211da90c12db9
## What changed
Instrument the spawned WebSocket startup prewarm task with a
`startup_prewarm` trace span. Set its OpenTelemetry name and attach the
session's `thread.id` so prewarm work can be associated with its session.
GitOrigin-RevId: f2d642d145686b883cd82be8a594bd2383470651
## Why
Clients could discover threads with paginated history, but could not page their
turns through `thread/turns/list`.
## What changed
- Route `thread/turns/list` for paginated threads through the thread store,
preserving cursors, sort direction, turn status, timing, errors, and the
requested `summary` or `notLoaded` item view.
- Reject the `full` turn item view for paginated threads and direct clients to
`thread/items/list` for complete item pagination.
- Return each item page entry with both its containing `turnId` and `item`, so
clients can regroup thread-wide pages into turns.
- Expose local thread-store turn and item pagination through the `ThreadStore`
implementation.
## Testing
Added app-server coverage for paging projected turns and items, item views,
cursors, turn metadata, and rejection of the unsupported `full` view.
GitOrigin-RevId: 9c4a127a9b4fa91313f0e87494c62fe2e3a06f19
## Why
The generic `yield_time_ms` description does not explain the Windows-specific
initial wait behavior or when callers should choose a shorter value.
## What changed
- Describe `yield_time_ms` on Windows as the maximum wait before a running
command returns a session ID, while commands that finish sooner still return
immediately.
- Recommend the 10-second default for ordinary commands and shorter waits for
intentionally long-lived or interactive processes.
- Keep the existing parameter description on other platforms.
## Testing
- Update the expected `exec_command` tool schema for both platform variants.
GitOrigin-RevId: 32f90b3a56604fcf1734cc81ba7ab2f0bec41643
## What changed
- Convert filtered MCP tool metadata into `CoreToolRuntime` instances before constructing the tool router.
- Pass those runtimes through the shared tool-planning path instead of maintaining separate direct and deferred MCP tool lists.
- Preserve MCP visibility, app policy filtering, and direct versus deferred exposure behavior.
GitOrigin-RevId: 67f7fbeecc2992318c859e038cd97938833911db
## What changed
- Add per-sandbox-mode permission messages to `ModelMessages` and preserve them when applying model configuration overrides.
- Use the selected catalog message in permission instructions, substituting `{{ network_access }}` with the active network policy. Fall back to the existing sandbox text when no override is provided, and allow an empty override to omit only the sandbox section.
- Apply catalog permission messages when a session starts and when its model changes.
## Testing
- Cover catalog deserialization, mode selection, network substitution, fallback and empty-message behavior, remote model catalogs, and model changes.
GitOrigin-RevId: cd5ed3aee3155dca0e7b2358c0f09bae73236856
## What changed
- Recommend simplifying prompts incrementally and evaluating each change while preserving product requirements.
- Document `text.verbosity` as the default response-detail control and clarify how prompts should specify task-specific content and tone.
- Clarify autonomy and approval boundaries for multi-step work.
- Narrow Programmatic Tool Calling guidance to bounded reduction workflows and call out validation of both `program_output` and the final assistant `message`.
GitOrigin-RevId: 2334a1cfba9ead53d2002a097920d1306bcae0d4
## What changed
- Accept a caller-provided `CodexAppsToolsCache` when constructing a `ThreadManager` and pass it through to the MCP manager.
- Add `ConnectorRuntimeManager::new_without_cache()` for an in-memory runtime that neither loads nor persists connector state on disk.
- Re-export `CodexAppsToolsCache` from `codex-core` and `codex-core-api` for embedders.
## Testing
- Verify that a cache-disabled connector runtime ignores existing disk state and publishes live tools without creating cache files.
GitOrigin-RevId: f52f4f27c425a8d84ad0d0987b30a6fc3c14d702
## What changed
- Check the source thread's `history_mode` before loading its history for
`thread/fork`.
- Return the existing `paginated_threads is not supported yet` method-not-found
error for paginated source threads.
## Testing
- Add an app-server integration test covering rejection of a paginated source
thread.
GitOrigin-RevId: 49b4b8aef7c9488eec93c24e831cea6382f8c995
## Why
Special filesystem subpaths can contain relative or foreign-platform path
spellings. Treating this wire text as a host-native path can reinterpret it.
## What changed
- Store special filesystem subpaths as strings in the core permissions model.
- Represent app-server subpaths as `LegacyAppPathString` and convert them at
the protocol boundary without changing their spelling.
- Add `LegacyAppPathString::from_string` for callers that already own legacy
app-server path text.
GitOrigin-RevId: 2214f03aa997fd0e6eb2ab60d425304bb4bd6132
## What changed
- Hide the working-status indicator when finalized assistant output is committed to the transcript, including single-line final answers.
- Restore the indicator when image generation begins during a running task after its streamed preamble has been flushed.
## Testing
- Add snapshot coverage for a single-line final answer and for image generation following a single-line preamble.
GitOrigin-RevId: e294a11cda3f0ce46f2bbed1be2a1380c9c73085
## Why
Stream consolidation can replace transcript cells while initial history replay is buffering capped display rows, leaving those rows stale and omitting the finalized output.
## What changed
- Discard buffered rows when required stream reflow occurs during capped initial replay and rebuild the capped tail from the canonical transcript after replay finishes.
- Schedule the rebuild through the resize reflow path so it waits for an active transcript overlay to close.
## Testing
Added coverage for consolidated Markdown output during capped initial replay, both with and without an active transcript overlay.
GitOrigin-RevId: 1c184ff561a50362daf11c1c629be59882f27a4a
## What changed
- Add `codex-agent-extension` with an `AgentRunner` that starts a resolved agent prompt in a thread forked from its parent.
- Propagate the invocation's trace context, select the configured execution environments, submit the initial prompt, and return the spawned thread and turn identifiers.
- Reject empty agent prompts and report when the owning thread manager is no longer available.
## Testing
- Add an integration test that verifies the agent runs in a forked thread, returns the started turn identifier, completes the turn, and sends the resolved prompt to the model.
GitOrigin-RevId: ffb805efabadf758969d6611a6867db6fa9f059a
## What changed
Allow API key sessions to send plugin-attributed `codex_plugin_used`,
`skill_invocation`, and `codex_mcp_tool_call_event` events. Continue to drop
events without a plugin ID and all other analytics events for these sessions.
## Testing
Add coverage that mixes plugin and non-plugin events and verifies that only the
three plugin-attributed event types are delivered without server-owned auth
fields.
GitOrigin-RevId: 1f47af5e9516723154f566ad032dba3ebd98abf9
## What changed
- Default `prompt_cache_key` to the session ID from response metadata instead of the thread ID.
- Preserve explicit prompt cache key overrides.
- Verify that API-key-authenticated root and subagent requests use the same session-based cache key even though they have different thread IDs.
GitOrigin-RevId: cb1bf21a3cb8fa2c5083c5c6afd4829ff2adcf1f
## Why
Exec-server JSON-RPC payloads can contain decimals, exponent notation, and
integers outside the native 64-bit range. Decoding these values should preserve
their original JSON representation without weakening the existing message
complexity limit.
## What changed
- Enable `serde_json` arbitrary-precision number and raw-value support for the
exec-server protocol.
- Decode serde's number and raw-value wrappers in the bounded JSON visitor,
charging nested raw values against the 256K-value limit.
- Select the JSON-RPC envelope variant from its fields after bounded decoding so
arbitrary-precision values survive message deserialization.
## Testing
Add coverage for exact arbitrary-precision number round trips and for enforcing
the value limit inside raw-value wrappers.
GitOrigin-RevId: a6e0a3fba6f88e4e8a6ff414832f078beb2d2a70
## What changed
- Return only each live agent's name and status from `list_agents`.
- Stop tracking the most recent task message in agent metadata.
GitOrigin-RevId: 91836fe8efad7ea6b7305f0b409081ec77419cfd
## What changed
Add `rawResponse/completed` and its `RawResponseCompletedNotification` payload
to the generated `ServerNotificationEnvelope` union, keeping the envelope type
aligned with the supported server notifications.
GitOrigin-RevId: 25a8b5e05a146a66ea65928eb6f0c3bd7b80f082
