## Why
Tool lifecycle observers need visibility into calls rejected before execution starts and timing that remains available on cancellation, independently of tracing settings.
## What changed
- Add `on_tool_dispatch` and `on_tool_timing` callbacks to `ToolLifecycleContributor`, with thread, turn, and call IDs for correlation.
- Report direct calls before readiness, dispatch waiting, and hooks. Report handler duration when dispatch completes or is dropped, excluding readiness and gate waiting and returning zero if the call never passes the gate.
- Expose code-mode host durations as a separate `HostOperation` timing boundary. Exclude nested code-mode calls from dispatch and handler timing observations to avoid double counting.
## Testing
Add tests for callback ordering on completed and rejected calls, timing on cancellation, zero handler duration before gate admission without tracing, and code-mode host timing without nested dispatch or handler timing observations.
GitOrigin-RevId: 97998b5094564c13cb2cfad089bb0a9d6615eb5c
## What changed
Add `model_messages.tools.indirect_description_prefixes` with selectors for exact namespaces and configured MCP server names. Apply the selected model's guidance to namespace documentation embedded in `CodeModeOnly` prompts, `ALL_TOOLS` descriptions, and namespaces returned by tool search.
Trim prefix values, treat empty values as no prefix, and reject conflicting selectors for the same namespace. Apply prefixes during rendering so model changes take effect without altering direct tool specifications, search ranking, or tool dispatch.
## Testing
Add unit tests for prefix normalization, namespace grouping, MCP server mapping, and conflicts. Add scenarios covering both Code Mode variants, clearing prefixes on model changes, unchanged tool dispatch, and conflict rejection before a model request.
GitOrigin-RevId: 0c7ce6d418c775945c2ccfdab3c605fa4d431b55
## What changed
- Add `tools.code_mode` catalog messages for `exec`, `wait`, deferred-tool guidance, and shared MCP TypeScript definitions.
- Render `{{ default_exec_yield_time_ms }}` and `{{ image_helper }}` in `exec` descriptions while preserving runtime tool declarations and the freeform grammar.
- Resolve `wait` descriptions and parameter schemas independently, falling back to bundled parameters for invalid or unsupported schemas.
- Use the active model's Code Mode messages, including after mid-turn model changes. Missing overrides retain bundled defaults; empty text overrides suppress the corresponding text.
## Testing
Add coverage for literal template substitution, runtime section preservation, independent `wait` overrides and fallback behavior, mid-turn model changes, and a yielded cell completing through `wait` with catalog messages.
GitOrigin-RevId: e542a0cd47a7138a414871192cee47790ed1214e
## Why
Guardian can lose a confirmed User Messaging message from its parent context when a `PostToolUse` hook blocks the result or is cancelled. The message has already been sent and remains relevant to interpreting the user's authorization.
## What changed
Capture confirmed message text after input-rewriting hooks in shared per-call state before awaiting post-tool processing. Attach that evidence to the response metadata even when dispatch returns a blocked or aborted result, while keeping post-tool hooks cancellable. Calls cancelled before delivery confirmation contribute no message evidence.
## Testing
Extend Guardian subagent authorization coverage to verify retained messages after post-tool hook blocking and cancellation, and their absence when cancellation precedes delivery confirmation.
GitOrigin-RevId: 9abb7959980eada20d0fb5c0b3a4a8669e4cd0a3
## Why
Code-mode responses show host duration without exposing time spent outside the host, including app-server waiting.
## What changed
- Add `features.code_mode.experimental_show_cell_overhead`, disabled by default, to show handler duration, host duration, and their difference in `exec` and `wait` response headers.
- Reuse the completed handler measurement from tool-call logging, excluding dispatch waiting, and capture it even when logging is disabled.
- Preserve the existing timing format when the option is disabled or host timing is unavailable, and preserve boolean feature toggles when merging nested code-mode configuration.
## Testing
Add coverage for timing headers, zero and missing host measurements, negative rounding differences, content preservation, and configuration merging. Extend app-server tests across gRPC and stdio to verify timing includes post-host elicitation waiting, and add a response snapshot for successful execution and a failed wait.
GitOrigin-RevId: 929b84784d9c6077cf2b5927a90bd2d5f3870078
## What changed
Represent images in `ContentItem` and `FunctionCallOutputContentItem` with `ImageReference::Inline`, flattened to preserve the existing `image_url` wire format. Update image producers and consumers and regenerate app-server schemas and SDK artifacts.
Preserve the Python SDK's `InputImageContentItem` and `InputImageFunctionCallOutputContentItem` class names during generation.
## Testing
Add a regression test for stable Python image class names and adapt existing image tests to the shared representation.
GitOrigin-RevId: c38a780ac3314c2ac2deb3afc1b93b94b6f93fec
## What changed
Include `sessionId` and the originating `windowId` alongside `threadId` and optional `itemId` in MCP request metadata. Retain the originating item and window for code-mode cells across waits and compaction, including the window when no matching history item is found.
## Testing
Extend metadata assertions for direct and nested MCP calls. Add a regression test verifying that a code-mode cell retains its original item and window IDs when it resumes after compaction.
GitOrigin-RevId: 5e3cf761f2bed4c16565334c94654496367325fd
## Why
A yielded code mode cell can emit a notification after the session switches models. Its output should retain the originating model's truncation budget instead of using the receiving model's budget.
## What changed
Carry the cell's output token limit, including the serialization allowance, through notification dispatch and attach it as `history_truncation_token_limit` metadata. Allow `inject_if_running` to accept response item envelopes so queued notifications preserve this metadata in history and rollouts.
## Testing
Add regression coverage for delayed notifications across a model switch, verifying per-cell truncation budgets, full notification text in raw events, and persisted rollout metadata. Add a task-completion test that checks notification budgets are preserved alongside queued user input.
GitOrigin-RevId: 28380a039a6d41ea324401fcf1f611d81a279477
## Why
A code-mode cell can outlive its turn. Nested tool calls dispatched during a later turn need the original step's tools and context, while approval handling must still work across that turn boundary.
## What changed
- Retain each cell's originating `StepContext` for nested tool dispatch and cancel queued calls whose context has been released.
- Keep MCP approval metadata available for the lifetime of the running invocation, keyed by server and call ID.
- Apply Guardian denial accounting and interruption to the active turn, including reviews from an earlier cell.
- Carry `turn_id` in legacy MCP begin/end events so history reconstruction and rollout migration assign late completions to their original turn. Preserve the current-turn fallback for older records without it.
## Testing
Add regression coverage for originating tool selection across turns, callback and metadata cleanup, delayed MCP elicitation review, Guardian interruption of the active turn, and late MCP completion attribution with legacy compatibility.
GitOrigin-RevId: 4533379aae6e4ecb4aabe5a25e4aca5d65385d76
## Why
A delegate bound at session creation cannot provide different callbacks for cells sharing that session. Each execution needs to retain its own delegate across yields and release it when the cell is cleaned up.
## What changed
- Pass `CodeModeSessionDelegate` to `execute` instead of session creation.
- Route tool calls, notifications, and cell closure callbacks through the execution's delegate in the in-process runtime and the gRPC and stdio transports.
- Retain delegates with pending executions and live cells, releasing them through closure and cancellation cleanup.
## Testing
Add coverage for distinct delegates across yielded cells, gRPC callbacks before cell admission, and delegate release after completion or abandoned execution cleanup. Update transport tests to verify callbacks reach the owning cell's delegate.
GitOrigin-RevId: 7469f52104b993e790a40c65fb800ad02b7fd606
## Why
Tool results can be ready before the sampling loop collects them, and code-mode cells can outlive a turn. Separate trace milestones make these timings visible and link nested calls to the turn that dispatches them.
## What changed
- Emit `codex.tool_call_received` and `codex.tool_result_ready` events for direct and code-mode calls, using identifiers and tool names without arguments or output. Record readiness before result encoding or collection, including ordinary errors and aborted responses, but excluding fatal errors.
- Assign nested call IDs at broker receipt and carry them through dispatch. Trace dispatch and interruptions caused by cancellation, cell closure, or abandonment.
- Add code-mode execute and wait handler spans with conversation, turn, call, and cell identifiers and `completed`, `failed`, or `interrupted` outcomes.
GitOrigin-RevId: b098a29d8ee030c823665e4b1031a0d67cc09502
## Why
After a model change within a turn, tools could still use the turn's original model capabilities, output limits, and telemetry attribution.
## What changed
- Build each step's tool plan from its captured model settings.
- Use those settings for image support and detail, command and MCP output truncation, extension metadata, and tool telemetry.
- Preserve the issuing step's model for delayed command completion events while allowing later `write_stdin` calls to use their own step's output limits.
## Testing
Add regression coverage for model changes within a turn, including tool replanning, Code Mode execution, image handling, MCP tool and resource output limits, extension settings, and delayed command completion.
GitOrigin-RevId: ef991598e89a846889baf770a0c9e12e56886d5d
## What changed
Introduce `ExecutedToolCalls` to own optional shared recorder state and centralize feature checks, call recording, and accepted result source lookup. Update session, tool runtime, MCP, and Code Mode callers to use this interface.
Move prompt attachment, retry handling, and metadata budgeting into the `request_metadata` module, preserving existing recording and request behavior.
## Testing
Add coverage for all session and turn feature combinations, lazy result source lookup, and unchanged prompts when recording is disabled. Retain the existing metadata budget and cell completeness tests in the new module.
GitOrigin-RevId: bc19f34317232b6599c7ec71c15db4a80123ff6f
## Why
Code-mode callbacks can resume outside the turn's local span ancestry, leaving tool dispatch traces without the thread identity inherited from a turn span.
## What changed
Record `thread.id` directly on the `dispatch_tool_call_with_code_mode_result` span using the session's thread ID.
## Testing
Add a regression test that dispatches a code-mode tool callback from a fresh task without a turn span and verifies that the dispatch span records the thread ID, tool name, and an `exec-` call ID with a valid UUID.
GitOrigin-RevId: a25a0d8fcaf1b6501473151c486fc9940065df91
## Why
Code-mode callbacks can outlive their initial request and run in separate tasks,
so their trace context must be preserved explicitly.
## What changed
- Preserve the execution context for code-mode callbacks and add spans for
nested tool invocations in the runtime and dispatch broker.
- Propagate each tool invocation span as the parent of its streamed gRPC
callback, falling back to the outer execution context when needed.
- Add a `codex.exec_server.process` span for each process lifetime, including
its process ID and completion result, without retaining the request span.
GitOrigin-RevId: a6059e34d895416f5517e51dad6ca0078355adca
## Why
Code mode wall time should measure the host operation itself, without including
client-side response delays or idle time between requests.
## What changed
- Measure each execute, wait, and terminate request in the code mode host.
- Carry the duration through the stdio and gRPC protocols and use it for
model-visible wall time.
- Emit a structured `codex.code_mode.host_timing` event correlated with the
conversation, turn, tool call, and cell.
## Testing
- Cover successful and failed execution timing, delayed response reads,
repeated waits, termination, and missing cells across stdio and gRPC.
- Verify timing survives protocol serialization and is reflected in app-server
model output and structured telemetry.
GitOrigin-RevId: d24af30c3fc5820521b4beba1f9970714dad6482
## Why
Planning tools for a candidate model must not change the selected model or its
Responses Lite tool inventory. Otherwise, preparing a fallback can overwrite
metadata before the current model's request is sent.
## What changed
- Build tool registries and visible specs from an explicit `ModelInfo`.
- Store the finalized tool mode, code-mode names, namespace inventory, and
terminal and child-management capabilities on `ToolRouter`.
- Publish namespace inventory only after selecting the step's router, and use
that router's mode when dispatching and recording tool calls.
## Testing
Added coverage for candidate-model isolation, fallback compaction metadata,
code-mode worker selection, normalized tool names, and router capabilities.
GitOrigin-RevId: 56df2de8f42c7e93a4c4dcd117720e40a0cca3db
## Why
Executed-tool metadata for a Code Mode cell can span its `exec` and `wait`
outputs. Consumers need to know whether that metadata contains the cell's full
tool call inventory or only a partial record.
## What changed
- Associate recorded tool calls with the originating Code Mode cell across
`exec` and `wait` outputs.
- Add a host-owned `tool_calls_complete` marker when recording finishes without
dropped or truncated calls. The marker describes inventory completeness, not
tool success.
- Preserve the cell metadata through replay and pruning, and clear the
completeness marker when prompt budgeting makes the record incomplete.
## Testing
Add coverage for multi-wait cells, retries, interruptions, recording limits,
argument truncation, prompt budgeting, and rejection of input-forged metadata.
GitOrigin-RevId: 377014edec06aab42aadab78b8b5dccc0a1c83ba
## What changed
- Add callback lifetimes to extension `ToolCall`, `ToolEnvironment`, turn-input context, and skill-read request types.
- Require extension tool executors to handle calls for any invocation lifetime and tie their returned futures to that lifetime.
- Update built-in extensions, adapters, and tests to use the lifetime-scoped APIs without retaining host-owned environment capabilities.
GitOrigin-RevId: aca9ce3a1b6870df550bc84ec9308cc65533f161
## What changed
- Inject W3C `traceparent` metadata into code-mode session and execution requests.
- Carry each execution's trace context on streamed tool callbacks so callback and nested-tool spans remain connected across the gRPC boundary.
- Preserve the relevant spans when execution, callback, and dispatch work moves into spawned tasks.
GitOrigin-RevId: 6482963cb9b2e2f0ea27f1345bba208b70a80d0b
## What changed
- Add the originating Responses item ID to MCP tool request metadata as
`_meta.itemId`.
- Preserve the initiating `exec` item ID across code-mode cells so nested MCP
calls carry the same origin.
## Testing
- Cover metadata generation for direct and code-mode MCP tool calls.
GitOrigin-RevId: 1161cac64f1cc3a6d4dac4368e63804bd9d2c417
## What changed
- Add the opt-in `code_mode_prewarm` feature to establish the code-mode host session during startup, before the first turn.
- Make in-progress host initialization cancellable so a stalled prewarm does not block shutdown.
- Share a failed connection attempt with concurrent callers instead of immediately starting another attempt.
## Testing
- Verify app-server contacts the configured host before the first turn and can shut down while that connection is stalled.
- Exercise shared remote-host behavior with prewarming enabled.
GitOrigin-RevId: 3cfde5509be7b4aba80112c36c5616b9a9b632c6
## Why
Thread settings can change while a turn is running, and delegated review or
compaction steps can select a different model. Each request needs a consistent
set of settings resolved against the model that will execute it.
## What changed
- Capture immutable resolved settings for each model step, including model
metadata, reasoning options, service tier, approvals, and personality.
- Keep in-flight steps on their captured settings while applying thread updates
to subsequent work.
- Resolve inherited reasoning summaries and filter service tiers against the
selected model and feature configuration for review and compaction steps.
## Testing
Added coverage for settings updates during paused turns, model changes,
previous-model compaction, review-model defaults, and service-tier filtering.
GitOrigin-RevId: fca4b3b305379f1bf26a5cac75ee0dd1747ac1f6
## Why
Analytics need to associate subagent activity with the top-level turn that
caused it, while avoiding a stale association when steering makes that origin
ambiguous.
## What changed
- Add `root_turn_id` to turn and tool-item analytics events.
- Read the root from live turn metadata when each event is built, reporting
`null` when the root is unknown or ambiguous.
- Propagate the metadata through control-tool and code-mode completion facts so
correlated and background tool events retain the correct root.
## Testing
- Cover root IDs for top-level turns, subagent turns, correlated code-mode
calls, and tool events emitted after turn completion.
- Verify that queued and subsequent events report `null` after the live root is
invalidated.
GitOrigin-RevId: 42fd76541c3c716cec74787d47b50adb2d22e6d2
## Why
Interrupting a turn should also stop code-mode work that the turn left running.
## What changed
- Add the disabled-by-default `code_mode_interrupt` feature.
- When an interrupted turn has the feature enabled, terminate all active code-mode cells and propagate cancellation through nested tool calls.
- Keep the reusable code-mode session alive so its stored state remains available to later turns.
## Testing
Add an integration test that interrupts a long-running nested tool, verifies that active and background cells are removed, and confirms that session state is preserved.
GitOrigin-RevId: 2b8634776a89b636318a39b9e9ad2eefc07cead8
## What changed
- Add `features.code_mode.default_exec_yield_time_ms`, defaulting to 30 seconds.
- Apply the configured default to code-mode `exec` calls that omit `yield_time_ms` and reflect it in the tool description.
- Remove the `code_mode_buffered_exec` compatibility feature flag.
## Testing
- Cover config loading and tool descriptions for both default and custom yield timeouts.
GitOrigin-RevId: 581ffc737b04721c7b5fc81260b1da1cf7d87523
## Why
MCP tool handlers and their Code Mode definitions are immutable for the lifetime
of a stable MCP binding, so rebuilding them for every sampling step repeats schema
construction unnecessarily.
## What changed
- Cache MCP handlers per session and reuse them while the same binding remains
active, invalidating the cache when the binding changes.
- Lazily cache schema-free Code Mode definitions on MCP handlers while continuing
to rebuild definitions for step-local tools whose specifications may change.
- Reapply current app enablement, tool policy, exposure, and plugin schema budgets
whenever cached handlers are registered for a step.
## Testing
- Cover handler and Code Mode definition reuse, binding invalidation, changing app
policy, stable continuation schemas, and step-local custom tool updates.
GitOrigin-RevId: 69c5011fb97c08580ac476e7fb2286b1e92e0359
## What changed
- Add the gated `unified_image_budget` feature for models that support original image detail or Responses Lite.
- Apply one 6,000-pixel, 10,000-patch preprocessing limit regardless of legacy image detail hints.
- Hide detail controls from `view_image` and code mode while continuing to accept existing hints for compatibility.
- Preserve detail-based resizing and tool contracts for unsupported models.
## Testing
- Cover unified resizing limits, Responses Lite, legacy detail hints, unsupported models, and `view_image` integration in code mode.
GitOrigin-RevId: cb07bff1669a96599fdfd076b3d9ec80f2b6fff7
## What changed
- Group top-level function and custom tool definitions into a single `functions` namespace for Responses Lite providers that support namespaced tools.
- Normalize missing, empty, and explicit `functions` namespaces to the same tool identity across registration, routing, lifecycle hooks, configuration, and tool search.
- Keep default tool names unprefixed in code mode, display output, and dispatch traces while preserving explicit non-default namespaces.
## Testing
- Add coverage for Responses Lite serialization, tool search results, namespace normalization and collision handling, routing, lifecycle events, and code-mode namespace policies.
GitOrigin-RevId: d48414005b5d22d39b11a198e19c47814d3a19f2
## What changed
- Add optional `defer_loading` support to freeform Responses API tool definitions.
- Omit the field when it is unset so existing eager tool definitions retain their wire shape.
## Testing
- Verify legacy freeform tool deserialization and eager and deferred serialization shapes.
GitOrigin-RevId: 50e8658a54ac5b6ae0c1dbfe65f7bb62efef561d
## What changed
- Add `codex-utils-audio` as a workspace crate for canonicalizing audio inputs
and estimating their token usage.
- Update `codex-core` to consume the new crate while preserving the existing
audio preparation tests.
GitOrigin-RevId: d719ecc08363ef52778aa37f3df0ca14f7778324
## What changed
- Emit dynamic tool-call analytics for code mode `exec` and `wait` calls, including duration and terminal status.
- Associate code-mode cells and their nested tool calls with the originating and subsequent model response IDs.
- Buffer correlated tool events until the next response is known, while flushing pending events when turns, threads, or the analytics queue close.
## Testing
- Add reducer coverage for correlating `exec`, `wait`, and nested tool events across turns.
- Add an app-server test that verifies production analytics include the cell and response IDs for a code-mode `exec` call.
GitOrigin-RevId: 8b34ab59c1ad68ebbd57f9c8a05f19ab04df48a4
## Why
Tool calls can outlive the sampling request that advertised them, so execution
must retain the finalized tool plan for that specific step.
## What changed
- Resolve tool runtimes, parallelism, cancellation behavior, argument diff
consumers, and dispatch from the `ToolRouter` stored in `StepContext`.
- Remove the separate router parameter from `ToolCallRuntime` and code-mode
worker construction so callers cannot supply a different router.
- Add a test helper for installing a router into a test `StepContext`.
GitOrigin-RevId: 9e476ddd228ca5b1c03ed87863618ba77de6200c
## What changed
- Move the V8 implementation into a dedicated `codex-code-mode-runtime` crate used by `codex-code-mode-host`, removing the embedded runtime fallback from the Codex process.
- Resolve the host executable from the active installation layout and check its availability before selecting tools.
- Fall back to direct tools with a one-time warning when optional code mode is unavailable. Keep `code_mode_only` and `disable_in_process_fallback` configurations fail-closed.
## Testing
- Cover host discovery for standalone and package layouts, including missing hosts and symlinks.
- Verify direct-tool fallback, one-time warnings, and fail-closed code-mode-only behavior.
GitOrigin-RevId: 5aa3c6f1db148b2231fc24089a2ee0e2b00dbddb
## What changed
- Add the opt-in `executed_tool_call_metadata` feature to record model-attempted direct and code-mode tool calls.
- Attach recorded names and arguments to the matching tool output in the next Responses request, preserving the metadata across sampling retries.
- Bound pending calls and argument sizes, and emit truncation metadata when limits are exceeded.
## Testing
- Cover disabled-by-default behavior, namespaced and nested calls, blocked or failed attempts, retry handling, and recorder limits.
GitOrigin-RevId: 5ce917b73e797b0f6904d46477b9d2cf8ea71bf2
## What changed
- Preserve `encrypted_function_args` on function calls so an empty list can mark plaintext collaboration arguments across request replay.
- Deliver `spawn_agent`, `send_message`, and `followup_task` payloads as structured plaintext agent messages when that marker is present; retain encrypted delivery otherwise.
- Redact plaintext collaboration arguments from tool and communication logs, and omit the metadata when sending requests to non-OpenAI providers.
## Testing
- Cover serialization of empty encrypted-argument metadata and plaintext versus encrypted subagent message delivery.
- Verify plaintext tool arguments are redacted and provider-specific metadata is removed from non-OpenAI requests.
GitOrigin-RevId: 64db98ff0b61a3af2f04ed609292363f2e2362a8
## What changed
- Add the experimental `code_mode_buffered_exec` feature.
- When enabled, default code-mode `exec` calls without an explicit
`yield_time_ms` to 30 seconds instead of 10 seconds.
- Report the effective default in the model-visible `exec` tool description.
## Testing
- Verify that the generated tool description advertises the 30-second default
when the feature is enabled.
GitOrigin-RevId: 0eea4d0105137f09050ce31aad18c25c79df6b2a
## Why
Audio attachments could be dropped from tool outputs and reconstructed thread history, while inline base64 payload size could overstate their model context cost.
## What changed
- Preserve remote and local audio attachments in legacy user-message events and thread history.
- Keep audio from function and custom tool outputs when the model supports it and the output fits the truncation budget; otherwise replace it with an omission marker.
- Estimate inline audio cost from its duration, with a serialized-size fallback for malformed or unsupported data URLs, so compaction and truncation use model-visible cost instead of raw base64 size.
- Replace audio in messages and tool outputs with an explanatory placeholder for models without audio input support.
## Testing
Add coverage for history round trips, dynamic tool responses, modality filtering, duration-based context estimates, output-budget truncation, and remote compaction.
GitOrigin-RevId: 5ed4628850a668f8383b7cd73db71b144cc02cd1
## Why
Not every Codex distribution currently includes the
`codex-code-mode-host` companion binary. Enabling the process-host
feature should not make code mode unavailable on those surfaces while
packaging support is being completed.
## What changed
- Fall back to an in-process code-mode session only when spawning the
companion binary returns `io::ErrorKind::NotFound`.
- Keep permission, handshake, timeout, and other host failures visible
instead of silently falling back.
- Store the provider's owned-process/in-process choice as one
enum-backed state so later sessions reuse the fallback decision.
- Preserve the underlying spawn `io::Error` while retaining the host
path in the displayed error.
- Update provider, `CodeModeService`, and end-to-end coverage to verify
successful fallback execution.
## Test plan
- `just test -p codex-code-mode`
- `just test -p codex-core missing_process_host`
## Why
When MCP tools are deferred behind `tool_search`, Code mode keeps them
callable but omits their individual declarations from the initial `exec`
description. The shared MCP `CallToolResult` types were derived only
from directly rendered tools, so deferring every MCP tool also removed
the common response contract that models need to interpret MCP results.
This restores that contract without undoing the context savings from
deferred tool definitions. This is a follow-up to #29486.
## What changed
- Track deferred Code-mode tool definitions separately from directly
rendered definitions.
- Render the shared MCP type preamble when either direct or deferred MCP
tools are available.
- Keep deferred tool declarations out of the initial prompt.
- Add unit and integration coverage for deferred MCP tools.
## Testing
- `just test -p codex-code-mode-protocol`
- `just test -p codex-core
code_mode_only_guides_all_tools_search_and_calls_deferred_app_tools`
## Why
Code-mode tool results could return to the model while an MCP
elicitation was still waiting for user input. This differed from
parallel tool calling and could let the model continue before the user
resolved the request.
We need one session-level view of outstanding elicitations so tool
runtimes can consistently hold results until every pending elicitation
is resolved.
## What changed
- Added a counted, session-owned ElicitationService with RAII
registrations.
- Registered both core-originated and server-originated MCP elicitations
with the service.
- Migrated out-of-band elicitation tracking and unified exec timeout
pausing to the shared service.
- Made code-mode functions.exec and functions.wait capture their runtime
result normally, then hold it before returning while an elicitation is
outstanding.
- Kept terminate: true immediate; only its result is held.
- Preserved model-visible wall time across the elicitation hold.
- Kept the behavior session-scoped, with concurrent elicitations holding
the pause until all registrations are released.
## Summary
- add the `code_mode_host` feature flag and select
`ProcessOwnedCodeModeSessionProvider` in `CodeModeService` when enabled
- initialize code-mode sessions lazily so a missing host reports a tool
error without failing thread startup
- resolve `codex-code-mode-host` beside the running Codex binary by
default while preserving `CODEX_CODE_MODE_HOST_PATH` as an override
- add unit and end-to-end coverage for host resolution and graceful
missing-host behavior
## Why
This wires the process-owned session client from #30112 into the core
service behind an opt-in rollout gate. Packaged Codex installations can
place the helper in the same `bin` directory as the main executable
without relying on `PATH`, while development and custom installations
can continue to override the helper path.
## Stack
- Depends on #30112
- Base branch: `cconger/process-owned-session-runtime-4-client`
## Validation
Build `codex` and `codex-code-mode-host`
`CODEX_CODE_MODE_HOST_PATH="$PWD/target/debug/codex-code-mode-host"
./target/debug/codex --enable code_mode_host`
Mechanical rename of CodeModeService => InProcessCodeModeSession
This already implements a CodeModeSession as its prime interface to
Core. The name was vestigial _and_ confusing af when embedded inside
core::tools::code_mode::CodeModeService
## Why
With deferred executors, an environment can become ready between two
sampling requests in the same turn. The model-visible environment
update, advertised tools, and eventual tool execution must all describe
the same request-time view.
Otherwise, a request built while only environment B is ready can
advertise a tool without an `environment_id`; if higher-priority
environment A becomes ready before execution, that call could silently
run in A instead.
This PR is stacked on #29527.
## Design
`run_turn` captures one `Arc<StepContext>` at each sampling-request
boundary. That step owns the request's `TurnContext` and environment
snapshot.
- World-state environment updates and tool planning borrow that same
step.
- `ToolCallRuntime` retains the `Arc` while asynchronous tool calls
execute.
- `ToolInvocation` carries the step to handlers; its temporary `turn`
compatibility field is derived from the same object.
- `ToolRouter` does not retain `StepContext`; it only uses it while
constructing the request's tool set.
- With `DeferredExecutor` disabled, step capture keeps using the
environments frozen at turn start.
Simply: every sampling request gets one consistent picture of its
environments, from what the model sees through where its tool calls run.
## What changed
- Build environment-dependent tool specs from the current request's
`StepContext`.
- Use that same step for unified exec, legacy shell, `apply_patch`,
`view_image`, and `request_permissions` execution.
- Hide environment-backed tools, including `request_permissions`, while
no environment is attached.
- Resolve legacy shell paths and metadata from the selected step
environment instead of the stale turn-start environment.
- Capture explicit steps at non-turn-loop boundaries such as compaction,
prompt debug, and startup prewarm.
- Reconcile prompt-debug history from the same step used to build its
tools.
## Follow-up
- Bind yielded code-mode cells to the tool runtime that created them, so
nested calls made after yielding continue to use the originating
request's `StepContext`.
## Test plan
- `just test -p codex-core
deferred_executor_updates_context_and_tools_after_startup`
- `just test -p codex-core
environment_count_controls_environment_backed_tools`
- `just test -p codex-core
build_prompt_input_includes_context_and_user_message`
## Description
This PR cuts Codex over from generic `ResponseItem.metadata` (introduced
here: https://github.com/openai/codex/pull/28355) to
`ResponseItem.internal_chat_message_metadata_passthrough`, which is the
blessed path and has strongly-typed keys.
For now we have to drop this MAv2 usage of `metadata`:
https://github.com/openai/codex/pull/28561 until we figure out where
that should live.
## Why
`ResponseItem` variants do not have a consistent internal ID shape: some
variants carry required IDs, some carry optional IDs, and some cannot
represent an ID at all. The existing fields also use inconsistent serde,
TypeScript, and JSON-schema annotations. A single enum-level access path
is needed before history recording can assign and retain IDs.
This PR establishes that internal model only. It intentionally does not
generate or serialize IDs; allocation and wire persistence are isolated
in the stacked follow-up.
## What changed
- Give every concrete `ResponseItem` variant an `Option<String>` ID
field.
- Apply the same internal-only annotations to every ID field:
`#[serde(default, skip_serializing)]`, `#[ts(skip)]`, and
`#[schemars(skip)]`.
- Add `ResponseItem::id()` and `ResponseItem::set_id()` as the shared
accessors.
- Preserve IDs when history items are rewritten for truncation.
- Adapt consumers that previously assumed reasoning and image-generation
IDs were required.
- Regenerate app-server schemas so the hidden fields are represented
consistently.
The serde catch-all `ResponseItem::Other` remains ID-less because it
must remain a unit variant.
## Test plan
- `cargo check --tests -p codex-core -p codex-api -p codex-rollout-trace
-p codex-image-generation-extension`
- `just test -p codex-protocol`
- `just test -p codex-app-server-protocol`
- `just test -p codex-api -p codex-rollout-trace -p
codex-image-generation-extension`
- `just test -p codex-core event_mapping`
## Why
`TurnContext` had accumulated dead fields and cached projections of
values already owned by its per-turn `Config` or `ModelInfo`. Keeping
both copies made ownership unclear and allowed artificial split-brain
states, such as a compatibility hash differing from the model metadata
it came from.
`Prompt` similarly carried a write-only personality after personality
selection had already been materialized into its base instructions.
This makes the canonical owner explicit: configuration-backed values
come from `config`, model-derived values come from `model_info`, and
prompts contain only data consumed by request construction.
## What changed
- Remove the unused `ghost_snapshot`, `codex_self_exe`, and
`thread_source` fields.
- Remove duplicate `comp_hash`, `truncation_policy`, `features`,
`shell_environment_policy`, `codex_linux_sandbox_exe`, `compact_prompt`,
and `tool_mode` fields.
- Read those values directly from `TurnContext::config` or
`TurnContext::model_info` at their consumers.
- Remove the write-only `Prompt::personality` field and its constructor
assignments.
- Preserve review-turn inheritance of the parent turn's shell policy,
Linux sandbox executable, and compact prompt through the review config.
## Testing
- `cargo check -p codex-core --tests`
## Description
This PR adds an optional `metadata` field to `ResponseItem` for
Responses API calls. Only mechanical plumbing, no actual values
populated and sent yet. Turns out just adding a new field to
`ResponseItem` has quite a large blast radius already.
This change is backwards compatible because `metadata` is optional and
omitted when absent, so existing response items and rollout history
without it still deserialize and requests that do not set it keep the
same wire shape. For provider compatibility, we strip out `metadata`
before non-OpenAI Responses requests so Azure and AWS Bedrock never see
this field.
My followup PR here will actually make use of it to start storing and
passing along `turn_id`: https://github.com/openai/codex/pull/28360
## What changed
- Added `ResponseItemMetadata` with optional `turn_id`, plus optional
`metadata` on Responses API item variants and inter-agent communication.
- Preserved item metadata through response-item rewrites such as
truncation, missing tool-output synthesis, compaction history
rebuilding, visible-history conversion, rollout/resume, and generated
app-server schemas/types.
- Strip item metadata from non-OpenAI Responses requests while
preserving it for OpenAI-shaped requests.
- Updated the mechanical fixture/test construction churn required by the
new optional field.