## Summary
The Responses API may or may not send back metadata on response items.
So when comparing request data to the last response, we should discard
it and only consider the content. This change results in a higher
success rate of incremental requests.
Notably, there is one subtle change to the behavior here - we are
now ignoring metadata when comparing the previous request items to the
new request items as well. This should be fine, since the metadata is
explicitly out of scope for the comparison, regardless of whether the
item is an item from the request or response.
## Testing
- [x] Added unit test coverage for both cases of metadata existence and
absence.
## Why
Codex telemetry pipeline needs a per-request TTFT value. The existing
`codex.turn_ttft` is recorded once per turn, so it cannot represent
later inference requests in the same turn and can miss the beginning of
hidden reasoning.
This restores the low-volume per-request signal proposed in
https://github.com/bk-nvidia/codex/pull/3 without bringing back
per-WebSocket-event TRACE logging.
## What changed
- start a timer when each mapped Responses stream begins
- latch the timer on the first `response.output_item.added`, including
an empty hidden-reasoning item
- attach `ttft_ms` to the existing `codex.sse_event` /
`response.completed` telemetry record
- cover the new completion field with an integration test
## Semantics
The value is per inference request, not per turn. It measures
mapped-stream-to-first-output-item latency, matching the
customer-proposed metric. For HTTP, the stream is already established
before timing begins, so request setup and response-header latency are
excluded.
`response.output_item.added` is a client-visible proxy for the start of
hidden reasoning; this does not claim access to the server's internal
first raw-token timestamp.
## Validation
- `just test -p codex-otel` (47 passed)
- `just test -p codex-core process_sse_emits_completed_telemetry` (1
passed after the final timer-placement change)
- attempted `just test -p codex-core`: 2,855 passed and 53 failed
because of unrelated local-environment failures (missing
`test_stdio_server` fixture binary, shell startup noise, and
timing-sensitive tests); the focused telemetry test passed in that run
as well
## Why
The Bedrock GPT-5.6 catalog advertises `max`, but Codex treated it as an
opaque custom effort. That made the reasoning picker render it as
lowercase `max` while known efforts use productized labels.
Making `max` a known effort aligns catalog data, parsing, and UI
presentation without changing the `max` wire value or persisted
representation.
## What changed
- Add first-class `ReasoningEffort::Max` parsing and serialization.
- Use the typed effort in the Bedrock catalog and render it as `Max` in
the TUI.
- Preserve forward-compatible custom-effort coverage with a genuinely
unknown `future` value.
### Before
<img width="559" height="124" alt="Screenshot 2026-06-28 at 12 08 47 PM"
src="https://github.com/user-attachments/assets/7c43cf4f-020b-4605-9239-0a9c97eb7364"
/>
### After
<img width="558" height="107" alt="Screenshot 2026-06-28 at 12 09 10 PM"
src="https://github.com/user-attachments/assets/b9cc5ded-c940-43b4-b024-bba25abe0a17"
/>
## Stack
This is PR 3 of the simplified HAI single-run-task stack:
- [#19047](https://github.com/openai/codex/pull/19047) Agent Identity
assertion and task-registration primitives, including the shared
run-task helper used by existing Agent Identity JWT auth.
- [#19049](https://github.com/openai/codex/pull/19049)
Disabled-by-default ChatGPT auth opt-in that provisions/reuses persisted
Agent Identity runtime auth and its single run task.
- [#19051](https://github.com/openai/codex/pull/19051) Run-scoped
provider auth that uses one backend-owned task id for first-party
inference and compaction requests.
[#19054](https://github.com/openai/codex/pull/19054) collapsed out of
the active stack because the simplified design no longer needs a
separate background/control-plane task helper.
## Summary
This PR moves Agent Identity usage into provider auth resolution. That
keeps `AgentAssertion` auth tied to first-party OpenAI provider requests
instead of applying a late session-wide override that could affect
local, custom, Bedrock, API-key, or external-bearer providers.
What changed:
- adds a small `ProviderAuthScope` struct carrying the run auth policy
and session source needed by provider-scoped auth resolution
- lets `Session` opt the existing `ModelClient` into `ChatGptAuth`
policy when `use_agent_identity` is enabled, without adding a second
model-client constructor
- resolves Agent Identity only for first-party OpenAI provider auth
paths
- uses the persisted run task id from the `AgentIdentityAuth` record to
build `AgentAssertion` auth for Responses requests
- routes shared request setup through scoped provider auth so unary
compact requests use the same run-task assertion path as inference turns
- keeps local/custom/Bedrock/env-key/external-bearer provider auth
unchanged
- lets missing run-task state surface through the existing model-request
error path instead of silently falling back to bearer auth
This PR intentionally does not create thread-scoped, target-scoped, or
background-scoped task identities. The run task is the only task Codex
registers in this POC shape.
## Testing
- `just test -p codex-model-provider`
- `just test -p codex-core client::tests::provider_auth_scope_uses`
- `just test -p codex-core remote_compact_uses_agent_identity_assertion`
## Why
Ultra should be one user-facing reasoning selection for work that
benefits from both maximum reasoning and proactive multi-agent
delegation. Without it, clients must coordinate maximum reasoning with
the experimental `multiAgentMode` setting, even though the inference
backend still expects its existing `max` effort value.
This change makes reasoning effort the source of truth: clients select
`ultra`, core derives proactive multi-agent behavior when the turn is
eligible for multi-agent V2, and inference requests continue to use the
backend-compatible `max` value.
## What changed
- Add `ultra` as a first-class reasoning effort and preserve
model-catalog ordering when exposing it to clients.
- Convert `ultra` to `max` at the inference request boundary, including
Responses HTTP/WebSocket requests, startup prewarm, compaction, and
memory summarization.
- Derive effective multi-agent mode per turn from effective reasoning
effort:
- eligible multi-agent V2 + `ultra` → `proactive`
- eligible multi-agent V2 + any other effort → `explicitRequestOnly`
- V1 or otherwise ineligible sessions → no multi-agent mode instruction
- Keep the derived effective mode in turn context history so successive
turns can emit a developer-message update only when the effective mode
changes.
- Remove selected multi-agent mode from core session configuration, turn
construction, thread settings, resume/fork restoration, and subagent
spawn plumbing. Subagents inherit reasoning effort and derive their own
effective mode.
- Retain the experimental app-server `multiAgentMode` fields for wire
compatibility while marking them deprecated. Request values are accepted
but ignored; compatibility response fields report `explicitRequestOnly`.
- Display Ultra in the TUI using the order supplied by `model/list`.
## Validation
- `just test -p codex-core ultra_reasoning_uses_max_for_requests`
- `just test -p codex-tui model_reasoning_selection_popup`
## Why
Authentication mode is a domain concept used by login, model selection,
telemetry, and transports. Keeping the canonical type in app-server
protocol forces those lower-level crates to depend on an unrelated wire
API.
## What changed
- Added canonical `codex_protocol::auth::AuthMode` domain values.
- Kept the app-server wire DTO unchanged and added an explicit app-side
conversion.
- Removed production app-server-protocol dependencies from login,
model-provider-info, models-manager, and otel call paths.
## Stack
This is PR 2 of 6, stacked on [PR
#29714](https://github.com/openai/codex/pull/29714). Review only the
delta from `codex/split-json-rpc-protocols`. Next: [PR
#29722](https://github.com/openai/codex/pull/29722).
## Validation
- Auth and login coverage passed in the focused protocol/domain test
run.
- App-server account and auth conversion coverage passed.
## Why
Work(TPP) threads can be launched from the Desktop app, but if they all
keep the Desktop app's default originator then downstream attribution
cannot distinguish local Work launches from cloud-backed Work launches.
`thread/start.serviceName` already carries that launch signal, while
`SessionMeta.originator` is the durable thread-level value that survives
resume and fork.
This change converts the Desktop Work service names into an effective
originator at thread creation time, persists that originator with the
thread, and keeps using it for later model requests and memory writes.
## What changed
- Map `CODEX_WORK_LOCAL` and `CODEX_WORK_CLOUD` service names to
per-thread originators, while preserving
`CODEX_INTERNAL_ORIGINATOR_OVERRIDE` as the highest-precedence override.
- Persist the effective originator in `SessionMeta.originator`, read it
back on resume/fork, and inherit the parent originator for subagent
spawns when there is no persisted session metadata.
- Handle truncated `SpawnAgentForkMode::LastNTurns` forks by falling
back to the live parent originator when the forked history no longer
includes `SessionMeta`.
- Thread the per-thread originator through Responses headers,
websocket/compaction request paths, thread-store creation, rollout
metadata, and memory stage-one telemetry.
## Verification
- `just test -p codex-core
agent::control::tests::spawn_thread_subagent_inherits_parent_originator_without_fork
agent::control::tests::spawn_thread_subagent_fork_last_n_turns_inherits_parent_originator_without_session_meta
thread_manager::tests::originator_override_precedes_service_name_remapping`
- `just test -p codex-core
agent::control::tests::resume_thread_subagent_restores_stored_metadata_and_effective_multi_agent_mode`
- `just test -p codex-memories-write`
- `just fix -p codex-core -p codex-memories-write`
- `git diff --check`
When using Responses Lite, we should all use `additional_tools` and a
developer item instead of the top level tools array & instructions
field. This keeps things 1-to-1.
Forced namespacing for _all_ tools will land in a following PR after
some coordination & fixes in Responses API (around collisions & return
items).
The goal is to eventually expand the scope of this to _all_ requests
from codex, but that will require larger coordination across providers &
slower rollout.
## Summary
NVIDIA asked to measure Fast mode usage and reasoning effort from Codex
CLI OTEL logs. Add the finalized `service_tier` and
`model_reasoning_effort` to the existing `codex.sse_event`
`response.completed` record.
This intentionally reuses the existing completion event and leaves
transport APIs and shared telemetry plumbing unchanged.
## Testing
- `cargo build -p codex-cli --bin codex`
- `just test -p codex-core responses_api_emits_api_request_event`
- End-to-end with the built CLI and a local OTLP/HTTP collector:
- Fast/high emitted `service_tier=priority` and
`model_reasoning_effort=high` with token usage.
- Standard/low omitted `service_tier` and emitted
`model_reasoning_effort=low` with token usage.
## Why
Amazon Bedrock returns a `401 Unauthorized` response containing
`Signature expired:` when an AWS credential, including a short-lived
`AWS_BEARER_TOKEN_BEDROCK`, has expired. Codex currently surfaces that
response as a generic `unexpected status` error, which does not explain
how to recover.
Environment-provided bearer tokens cannot be refreshed automatically, so
the error should direct users to refresh their AWS credentials or
replace or remove the environment token and restart Codex. This
classification belongs to the Amazon Bedrock provider so similar
responses from other providers retain their existing behavior.
## What changed
- Add a synchronous `ModelProvider::map_api_error` hook that defaults to
the existing provider-neutral API error mapping, and route model
request, stream, WebSocket, and terminal unauthorized errors through the
active provider.
- Override the hook for Amazon Bedrock. After preserving the structured
status, body, URL, and request metadata, recognize `401` responses
containing `Signature expired:` and attach actionable credential
guidance.
- Keep `codex-protocol` provider-neutral by representing the guidance as
an optional `user_message`. Error rendering prefers this message while
continuing to append the URL, request ID, Cloudflare ray, and
authorization diagnostics.
- Add model-provider coverage for expired signatures and negative cases,
core coverage for provider dispatch after unauthorized recovery, and a
TUI snapshot for the rendered error.
## Testing
Tested with a real request with expired bedrock key:
<img width="962" height="126" alt="Screenshot 2026-06-22 at 3 56 51 PM"
src="https://github.com/user-attachments/assets/7e21cc7c-798e-4662-8467-7f304a2f2b59"
/>
## 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
Client-created response items enter history without IDs, so their
identity is lost across rollout persistence and resume. IDs should be
assigned once at the history-recording boundary, while IDs returned by
the server must remain unchanged.
The Responses API validates item IDs using type-specific prefixes.
Locally generated IDs therefore use the matching prefix plus a
hyphenated UUIDv7, keeping them valid while distinguishable from
server-generated IDs. Because this changes persisted history and
provider request shapes, the behavior is opt-in behind the
under-development `item_ids` feature. Compaction triggers remain request
controls whose API shape does not accept an ID.
## What changed
- Register the disabled-by-default `item_ids` feature and expose it in
`config.schema.json`.
- Make supported optional `ResponseItem` IDs serializable and expose
them in the generated app-server schemas.
- When `item_ids` is enabled, assign an ID during conversation-history
preparation if an item has no ID.
- Generate type-prefixed, hyphenated UUIDv7 IDs using the Responses API
item conventions.
- Preserve existing server IDs without rewriting them.
- Persist assigned IDs in rollouts and include them in subsequent
Responses requests.
- Remove the unsupported ID field from `CompactionTrigger` and document
why it has no ID.
- Add integration coverage for enabled ID persistence, preservation of
server IDs, and omission of generated IDs while the feature is disabled.
`prepare_conversation_items_for_history` is the single response-item ID
allocation boundary.
## Test plan
- `just test -p codex-features`
- `just test -p codex-core
response_item_ids_persist_across_resume_and_preserve_server_ids`
- `just test -p codex-core
non_openai_responses_requests_omit_item_turn_metadata`
- `just test -p codex-core
resize_all_images_prepares_failures_before_history_insertion`
- `just test -p codex-protocol`
- `just test -p codex-app-server-protocol`
- `just test -p codex-api azure_default_store_attaches_ids_and_headers`
## Summary
- Remove the realtime `architecture` selector from core protocol,
app-server protocol, config parsing, generated schemas, and callers.
- Always create WebRTC realtime calls with the AVAS query params:
`intent=quicksilver&architecture=avas`.
- Keep direct websocket realtime behavior on the existing config/default
path, while WebRTC starts without an explicit version now default to
realtime v1 because AVAS requires v1.
## Notes
- WebRTC realtime now means AVAS. If a caller explicitly asks to start
WebRTC with realtime v2, Codex rejects that request because the AVAS
WebRTC path only supports realtime v1. Websocket realtime is separate
and can still use realtime v2.
- The old `[realtime] architecture = "realtimeapi" | "avas"` config knob
is removed. Local configs that still set it will need to delete that
line.
- Some app-server tests that were only trying to exercise realtime v2
protocol behavior now use websocket transport, because WebRTC is
intentionally locked to AVAS/v1. Separate WebRTC tests cover the AVAS
query params, v1 startup, SDP flow, and sideband join.
## Validation
- Merged fresh `origin/main` at `83e6a786a2`.
- `just fmt`
- `just write-config-schema`
- `just write-app-server-schema`
- `git diff --check`
- `just test -p codex-api -p codex-core -p codex-app-server-protocol -p
codex-app-server realtime` (176 passed)
- `just test -p codex-protocol -p codex-config` (413 passed)
## 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.
## Why
WebSocket continuations only send the new part of a request. Checking
whether a request could be continued was cloning the full previous
request, the current request, and their input history.
For long conversations or large tool lists, that meant copying several
request-sized values on every continuation.
## What changed
- compare the request settings by reference
- check the previous input and server response as borrowed prefixes
- allocate only the new input items that will be sent
The reuse rules stay the same, including ignoring `client_metadata` for
this check.
The comparison is still `O(n)`, but it removes several `O(n)`
allocations and copies. Temporary memory no longer grows by multiple
full request sizes for each continuation.
## Performance
Local rollout traces show continuation checks on turns around 260k input
tokens. Before this change the reuse gate cloned the previous request,
the current request, and the previous input history before deciding
whether it could continue incrementally. After this change it borrows
those structures and allocates only the incremental tail. For large
continuations with a small delta, that removes roughly three
request-sized copies from the hot path and reduces temporary memory from
multiple full request sizes to just the new tail.
## Validation
- `just test -p codex-core
responses_websocket_v2_creates_with_previous_response_id_on_prefix`
- `just test -p codex-core
responses_websocket_v2_creates_without_previous_response_id_when_non_input_fields_change`
## Context
Inline compaction is part of the active logical turn. Compact requests
and the sampling requests around them should use the same turn state,
including when compaction is the first request to establish it.
## Change
Pass the turn-scoped `OnceLock` directly to inline v1 compaction so
`/responses/compact` includes an established value in the existing HTTP
header. Capture `x-codex-turn-state` from the compact response into that
same lock, allowing pre-turn compact to establish the value that
subsequent sampling reuses.
V2 compact already uses the normal Responses HTTP/WebSocket path and
continues to share the same `OnceLock` without separate plumbing. The
first returned value wins for the logical turn.
## Test plan
Integration coverage verifies that:
- pre-turn v1 compact can establish state for the first sampling request
- inline v1 compact receives established state over HTTP
- inline v2 compact reuses established state over HTTP
- inline v2 compact reuses established state over WebSocket
CI validates the full change.
## Context
Turn state is scoped to one logical turn, but the WebSocket path
currently exchanges it through upgrade headers, which are scoped to the
physical connection. A connection may be reused across turns, so its
handshake cannot represent the turn lifecycle reliably.
## Change
Exchange turn state on each WebSocket response request instead:
- send an established value in `response.create.client_metadata`
- read the returned value from the existing `response.metadata` event
- retain the first value in the turn-scoped `ModelClientSession`
`OnceLock`
- start the next logical turn without state, even when it reuses the
same WebSocket connection
This gives WebSocket requests the same first-value-wins contract as the
existing HTTP path.
## Test plan
Integration coverage verifies that:
- WebSocket replays returned state on same-turn follow-ups
- later response metadata does not replace the first value
- state resets at the logical turn boundary without requiring a
reconnect
CI validates the full change.
## Stack
This is 1/2. #28002 builds on this request-scoped transport to carry
established state through compact requests.
## Summary
Adds a `RealtimeConversationArchitecture` option for realtime
conversation startup, with `realtimeapi` as the default and `avas` as an
opt-in architecture.
The AVAS path is limited to realtime v1 conversational WebRTC starts,
and WebRTC call creation appends `intent=quicksilver&architecture=avas`
to `/v1/realtime/calls`. The existing sideband websocket still joins by
`call_id`.
This also exposes the per-session architecture override through
app-server v2 `thread/realtime/start` params and updates the config
schema for `[realtime].architecture`.
## Validation
- `just fmt`
- `just write-config-schema`
- `just test -p codex-api sends_avas_session_call_query_params`
- `just test -p codex-core -E
'test(~conversation_webrtc_start_uses_avas_architecture_query)'`
- `just test -p codex-core -E 'test(realtime_loads_from_config_toml)'`
- `just test -p codex-app-server-protocol -E
'test(~serialize_thread_realtime_start) |
test(generated_ts_optional_nullable_fields_only_in_params)'`
- `just test -p codex-app-server -E
'test(realtime_webrtc_start_emits_sdp_notification)'`
## What
Introduce a `CodexResponsesMetadata` struct that defines all the core
metadata we send to Responses API. Example fields are `thread_id`,
`turn_id`, `window_id`, etc.
Going forward, `client_metadata["x-codex-turn-metadata"]` will be the
canonical way Codex sends metadata to Responses API across both HTTP and
websocket transports.
For now, we continue to emit the existing top-level HTTP headers and
top-level `client_metadata` fields from the same
`CodexResponsesMetadata` struct for compatibility reasons.
Also, app-server clients who specify additional
`responsesapi_client_metadata` via `turn/start` and `turn/steer` will
have those fields merged into
`client_metadata["x-codex-turn-metadata"]`, but cannot override the
reserved fields that core uses (i.e. the fields in
`CodexResponsesMetadata`).
## Why
Responses API request instrumentation is the source of truth for
downstream Codex analytics that join requests by Codex IDs such as
session, thread, turn, and context window. Before this change, those
values were assembled through several request-specific paths: HTTP
request bodies, websocket handshake headers, websocket `response.create`
payloads, compaction requests, and the rich `x-codex-turn-metadata`
envelope all had their own wiring.
That made metadata propagation easy to drift across API-key/direct
Responses API requests, ChatGPT-auth/proxied requests, websocket
requests, and compaction requests. It also made additions like
`window_id` error-prone because a field could be added to one transport
projection but missed in another.
## What changed
- Added `CodexResponsesMetadata` as the core-owned snapshot for Codex
metadata sent to ResponsesAPI.
- Render `client_metadata["x-codex-turn-metadata"]`, flat
`client_metadata` projections, and direct compatibility headers from
that same snapshot.
- Include the known Codex-owned fields in the turn metadata blob,
including installation/session/thread/turn/window IDs, request kind,
lineage, sandbox/workspace metadata, timing, and compaction details.
- Treat app-server `responsesapi_client_metadata` as enrichment for the
Codex turn metadata blob while preventing those extras from overriding
Codex-owned fields.
- Use the same metadata path for normal turns, websocket prewarm, local
compaction, remote v1 compaction, and remote v2 compaction.
- Keep websocket connection-only preconnect metadata separate so
handshakes carry compatibility identity headers without inventing a fake
turn metadata blob.
## Verification
- `cargo check -p codex-core`
- `just fix -p codex-core`
## Summary
- Strip image `detail` fields from every Responses Lite request.
- Apply stripping to message images and function/custom tool-output
images.
- Transform only the formatted request copy without mutating stored
history.
- Preserve image URLs byte-for-byte, including HTTP(S) URLs, without
downloading, validating, or resizing them.
- Preserve all image `detail` fields for non-Responses-Lite models.
## Motivation
Responses Lite does not support image `detail` tags, so Codex must omit
them whenever `model_info.use_responses_lite` is enabled. This transport
requirement is independent of the `resize_all_images` feature.
Stored history retains the original detail values. This keeps
request-specific formatting isolated from conversation state and
preserves the information for local image preparation and
non-Responses-Lite requests.
#### [git stack](https://github.com/magus/git-stack-cli)
- ✅ `1` https://github.com/openai/codex/pull/27245
- ✅ `2` https://github.com/openai/codex/pull/27247
- 👉 `3` https://github.com/openai/codex/pull/27246
- ⏳ `4` https://github.com/openai/codex/pull/27266
## Why
Codex needs to manage Amazon Bedrock API key credentials through the
existing auth lifecycle instead of introducing a separate auth manager
or provider-specific credential file. Treating Bedrock API key login as
a primary auth mode gives it the same persistence, keyring, reload, and
logout behavior as the existing OpenAI API key and ChatGPT modes.
The credential is valid only for the `amazon-bedrock` model provider.
OpenAI-compatible providers must reject this auth mode rather than
treating the Bedrock key as an OpenAI bearer token.
## What changed
- Added `bedrockApiKey` as an app-server `AuthMode` and
`CodexAuth::BedrockApiKey` as a primary `AuthManager` mode.
- Added `BedrockApiKeyAuth`, containing the API key and AWS region, to
the existing `AuthDotJson` payload stored in `$CODEX_HOME/auth.json` or
the configured keyring backend.
- Added `login_with_bedrock_api_key(...)`, parallel to
`login_with_api_key(...)`, which replaces the current stored login with
Bedrock credentials.
- Reused generic auth reload and logout behavior instead of adding a
Bedrock-specific auth manager or logout path.
- Updated login restrictions, status reporting, diagnostics, telemetry
classification, generated app-server schemas, and auth fixtures for the
new mode.
- Added explicit errors when Bedrock API key auth is selected with an
OpenAI-compatible model provider.
This PR establishes managed storage and auth-mode behavior. Routing the
managed key and region into Amazon Bedrock requests will be in follow-up
PRs.
## Why
Compaction window identity is part of session history, not model-client
transport state. Persisting it with the compacted rollout item lets
resumed threads continue from the reconstructed window without keeping
mutable window state on `ModelClient`.
## What changed
- Added `window_id` to `CompactedItem` and stamp it when
`replace_compacted_history` installs compacted history.
- Moved auto-compact window id ownership into `AutoCompactWindow` /
`SessionState`; `ModelClient` now receives the request window id from
callers instead of storing it.
- Returned `window_id` from rollout reconstruction for resume.
Reconstruction uses the newest surviving compacted item's stored
`window_id` when present, and falls back to the legacy compacted-item
count when it is absent.
- Kept fork startup at the fresh default window id and updated direct
model-client tests to pass explicit test window ids.
## Validation
- `cargo check -p codex-core --tests`
## Summary
- Keep the existing `x-codex-window-id` HTTP header unchanged.
- Also send the same window ID in Responses `client_metadata`, allowing
supported backend paths to surface it as
`x-client-meta-x-codex-window-id`.
- Cover normal HTTP Responses and remote compaction v2 requests without
changing window generation or compaction behavior.
## Why
In the `2026-06-06T23` production hour, all 28,729 HTTP compaction
requests had `window_id` in `x-codex-turn-metadata`, but only 73
retained the direct `x-codex-window-id` header. The request-body
`client_metadata` path is already used for installation ID and is
preserved through supported Responses API paths.
This is additive metadata only. It does not change the direct header,
request count, model input, compaction routing, window generation, or
user response behavior.
Legacy `/v1/responses/compact` is intentionally unchanged. Its current
server-side `CompressBody` schema does not accept `client_metadata` and
rejects unknown fields, so supporting that path requires a backend
schema change before the Codex client can safely send this field.
## Validation
- Current head: `219baef3c`, rebased onto `origin/main` at `26d932983`.
- The post-rebase diff remains limited to the original five files (`22`
insertions, `6` deletions); the legacy experiment remains fully
reverted.
- `just test -p codex-core
responses_stream_includes_subagent_header_on_review`: passed; validates
normal HTTP Responses metadata.
- `just test -p codex-core
remote_compact_v2_reuses_compaction_trigger_for_followups`: passed;
validates remote compaction v2.
- `just test -p codex-core
remote_manual_compact_chatgpt_auth_reuses_service_tier_and_prompt_cache_key`:
passed; validates that legacy compact keeps its accepted payload shape.
- `just test -p codex-core
remote_manual_compact_api_auth_omits_service_tier_and_reuses_prompt_cache_key`:
passed; validates the legacy API-key payload as well.
- `just fmt`: passed; an unrelated root `justfile` rewrite produced by
the formatter was discarded.
- `git diff --check origin/main...HEAD`: passed.
The focused server pytest could not start in the local monorepo
environment because test setup is missing the `dotenv` module. Server
source and tests explicitly show that `CompressBody` omits
`client_metadata` and `/v1/responses/compact` returns HTTP 400 for
unknown body fields.
## Summary
- send `X-OpenAI-Internal-Codex-Responses-Lite: true` on HTTP Responses
requests and WebSocket upgrade requests when model metadata enables
Responses Lite
- use client metadata when sending it over the websocket
This PR is stacked on #26490.
## Why
The Responses Lite marker is request-scoped for HTTP but
connection-scoped for Responses-over-WebSocket because it is carried on
the upgrade request. Reusing a cached socket opened for the opposite
mode would therefore send the wrong transport contract.
## Validation
- `just test -p codex-core responses_lite`
- `just test -p codex-core
responses_websocket_reconnects_when_responses_lite_mode_changes`
- `just fix -p codex-core`
- `just fmt`
## Summary
- add v2 personal access token support for `codex login
--with-access-token` and `CODEX_ACCESS_TOKEN`
- classify opaque `at-` tokens separately from legacy Agent Identity
JWTs
- hydrate required ChatGPT account metadata through AuthAPI
`/v1/user-auth-credential/whoami`
- use PATs directly as bearer tokens while preserving existing ChatGPT
account surfaces
- expose PAT-backed auth as the explicit `personalAccessToken`
app-server auth mode
## Implementation
PAT auth is intentionally small and stateless. Loading a PAT performs
one AuthAPI metadata request, stores the hydrated metadata in the
in-memory auth object, and redacts the secret from debug output. Legacy
Agent Identity JWT handling remains unchanged. The shared access-token
classifier lives in a private neutral module because it dispatches
between both credential types.
PAT hydration fails closed when AuthAPI omits any required metadata,
including email. Hydrated metadata is intentionally not persisted:
startup performs a live `whoami` preflight so revoked tokens or changed
account metadata are not accepted from a stale cache.
## Workspace restriction scope
This change intentionally does **not** apply
`forced_chatgpt_workspace_id` to PAT authentication. The setting is a
client-side config guardrail, not an authorization boundary, and PAT
does not currently require workspace-ID parity. The PAT login and
`CODEX_ACCESS_TOKEN` paths therefore validate through AuthAPI without
threading workspace-restriction state through access-token loading.
Existing workspace checks for non-PAT auth remain on their established
paths.
## App-server compatibility
The public app-server `AuthMode` is shared across v1 and v2, and
PAT-backed auth reports `personalAccessToken` through both APIs.
Following human review, this intentionally removes the temporary v1
compatibility mapping that reported PATs as `chatgpt`; the deprecated v1
API is kept in parity with v2 rather than maintaining a separate closed
enum. Clients with exhaustive auth-mode handling in either API version
must add the new case and should generally treat it as ChatGPT-backed
unless they need PAT-specific behavior.
The v1 auth-status response still omits the raw PAT when `includeToken`
is requested because that response cannot carry the account metadata
needed to reuse the credential safely. Persisted PAT auth also omits the
new enum value so older Codex builds can deserialize `auth.json` and
infer PAT auth from the credential field after a rollback.
## Validation
Latest review-fix validation:
- `CARGO_INCREMENTAL=0 just test -p codex-login` (126 passed)
- `CARGO_INCREMENTAL=0 just test -p codex-cli` (263 passed)
- `CARGO_INCREMENTAL=0 just test -p codex-cli
stored_auth_validation_handles_personal_access_token`
- `CARGO_INCREMENTAL=0 just test -p codex-app-server-protocol` (226
passed)
- `CARGO_INCREMENTAL=0 just test -p codex-models-manager
refresh_available_models_uses_remote_only_catalog_for_chatgpt_auth`
- `CARGO_INCREMENTAL=0 just test -p codex-tui
existing_non_oauth_chatgpt_login_counts_as_signed_in`
- `CARGO_INCREMENTAL=0 just fix -p codex-login -p
codex-app-server-protocol -p codex-models-manager -p codex-tui -p
codex-cli`
- `just fmt`
- `git diff --check`
The broader `codex-tui` suite previously compiled and ran 2,834 tests.
Three unrelated environment-sensitive guardian/IDE-socket tests failed
after retries; the PAT-relevant TUI coverage passed.
## Summary
- add a defaulted `ModelInfo.use_responses_lite` catalog field
- support serializing `reasoning.context` while preserving the existing
effort and summary path
- has not been turned on for any models yet
I've added an override to parallel tools if responses_lite is on. I've
also forced persistent reasoning when using responses_lite. It would be
ideal if we could centralize all the responses_lite plumbing, but I
think this is best for now to keep the plumbing & diffs small.
## Testing
- `cargo test -p codex-protocol
model_info_defaults_availability_nux_to_none_when_omitted`
- `RUST_MIN_STACK=8388608 cargo test -p codex-core
responses_lite_sets_all_turns_context_and_disables_parallel_tool_calls`
- `RUST_MIN_STACK=8388608 cargo test -p codex-core
configured_reasoning_summary_is_sent`
- `cargo check -p codex-core --tests`
- `RUST_MIN_STACK=8388608 cargo clippy -p codex-core --tests` (passes
with pre-existing warnings in `codex-code-mode` and
`codex-core-plugins`)
## Summary
- accept non-empty model-defined reasoning effort values while
preserving built-in effort behavior
- propagate the non-Copy effort type through core, app-server, TUI,
telemetry, and persistence call sites
- preserve string wire encoding and expose an open-string schema for
clients
- update model selection and shortcut behavior for model-advertised
effort values
## Root cause
`ReasoningEffort` gained a string-backed custom variant, so it could no
longer implement `Copy` or rely on derived closed-enum serialization.
Existing consumers still moved effort values from shared references and
assumed a fixed built-in value set.
## Validation
- `just fmt`
- Local tests and compilation were not run per request; relying on CI.
## Why
The Responses websocket client no longer needs to send a follow-up
`response.processed` request after a turn response has already been
recorded. Keeping that extra acknowledgement path adds feature-gated
control flow and a second websocket request shape that no longer carries
useful behavior.
## What Changed
- Removed the `response.processed` websocket request type and sender.
- Removed the `responses_websocket_response_processed` feature flag and
schema entry.
- Removed turn and remote-compaction plumbing that only tracked response
IDs to send the acknowledgement.
- Removed tests that existed solely to cover the deleted feature path.
## Validation
- `just fix -p codex-core -p codex-api -p codex-features`
## Why
This PR
https://github.com/openai/codex/pull/24161#discussion_r3325692763
revealed a subagent data modeling issue, where we overloaded
`forked_from_id` to also mean `parent_thread_id`. That's incorrect since
guardian and review subagents can be a subagent and NOT fork the main
thread's history.
The solution here is to explicitly store a new `parent_thread_id` on
`SessionMeta`, alongside `forked_from_id` which already exists. While
we're at it, also expose it in the app-server protocol on the `Thread`
object.
A thread->subagent relationship and a fork of thread history are
orthogonal concepts.
## What Changed
- Added top-level `parent_thread_id` persistence on `SessionMeta` and
runtime/session plumbing through `SessionConfiguredEvent`,
`CodexSpawnArgs`, `SessionConfiguration`, `ThreadConfigSnapshot`,
`TurnContext`, and `ModelClient`.
- Made turn metadata, request headers, analytics, and subagent-start
events read the separate runtime/top-level parent field instead of
deriving general parent lineage from `SessionSource` or
`forked_from_thread_id`.
- Passed parent lineage separately at delegated subagent, review,
guardian, agent-job, and multi-agent spawn construction sites;
copied-history fork lineage remains derived only from `InitialHistory`.
- Persisted and exposed parent lineage through rollout/thread-store
projections and app-server v2 `Thread.parentThreadId`.
- Updated app-server README text and regenerated app-server schema
fixtures for the additive `parentThreadId` response field.
## Why
We recently added `forked_from_thread_id` which lets us trace where a
thread's _context_ comes from, but we also want to understand subagent
lineage (e.g. which parent thread spawned this subagent? what kind of
subagent is it?) which is orthogonal.
This PR adds `parent_thread_id` and `subagent_kind` to the
`x-codex-turn-metadata` header sent to ResponsesAPI.
## What changed
- Adds `parent_thread_id` and `subagent_kind` to core-owned
`x-codex-turn-metadata`.
- Restores persisted `SessionSource` and `ThreadSource` from resumed
session metadata so cold-resumed subagent threads keep their lineage on
later Responses API requests.
- Centralizes parent-thread extraction on `SessionSource` /
`SubAgentSource` and reuses it in the Responses client, analytics, agent
control, and state parsing paths.
- Extends reserved-key, git-enrichment, thread-spawn, and app-server v2
metadata coverage for the new lineage fields.
## Verification
- Not run locally per request.
- Added focused coverage in `core/src/turn_metadata_tests.rs` and
`app-server/tests/suite/v2/client_metadata.rs`.
Split from the Guardian prompt cache key change. This PR only updates
codex-rs/core/src/client.rs. Validation was not run per request; this
branch is expected to rely on the companion split PRs.
## Summary
- Add `request_kind` values for foreground turn, startup prewarm,
compaction, and detached memory model requests.
- Attach compaction dispatch metadata to local Responses, legacy
`/v1/responses/compact`, and remote v2 compact requests.
- Add the existing logical context-window identifier as `window_id` on
turn-owned model request metadata.
- Keep identity fields optional for detached memory requests, while
still emitting `request_kind="memory"` in non-git/no-sandbox workspaces.
## Root Cause
`x-codex-turn-metadata` has more than one producer. Foreground turns and
compaction requests own a real turn and should carry that turn identity.
Detached memory stage-one requests do not own a foreground turn, so
absent identity fields are valid rather than missing data. Startup
websocket prewarm is also a model request, but it has `generate=false`
and must not be counted as a foreground turn.
`thread_source` or session source identifies where a thread came from
(for example review, guardian, or another subagent). `request_kind`
identifies what the current outbound model request is doing (`turn`,
`prewarm`, `compaction`, or `memory`). A review or guardian thread can
issue either a normal turn request or a compaction request, so source
cannot replace request kind.
## Behavior / Impact
- Ordinary foreground requests send `request_kind="turn"`, their real
identity fields, and `window_id="<thread_id>:<window_generation>"`.
- Startup websocket warmup requests send `request_kind="prewarm"` so
they are not counted as foreground turns.
- Compaction requests send `request_kind="compaction"`, their real
owning turn identity, the existing `window_id`, and
`compaction.{trigger,reason,implementation,phase,strategy}`.
- Detached memory stage-one requests send `request_kind="memory"`
without `session_id`, `thread_id`, `turn_id`, or `window_id`; when no
workspace metadata exists, the kind-only header is still emitted.
- `session_id`, `thread_id`, `turn_id`, and `window_id` remain optional
in the header schema because detached memory requests do not own a
foreground turn or context window.
- `window_id` is not a new ID system: it is copied from the already-sent
`x-codex-window-id` / WS client metadata value at model-request dispatch
time.
- Existing `x-codex-window-id` HTTP/WS emission, value format,
generation advancement, resume behavior, and fork reset behavior are
unchanged.
- `request_kind`, `window_id`, and upstream turn-owned identity fields
remain schema-owned; input `responsesapi_client_metadata` cannot replace
their canonical values.
- No table, DAG, export, app-server API, or MCP `_meta` schema changes
are included.
A compaction attempt stopped by a pre-compact hook issues no model
request and therefore has no request header; its outcome remains in
analytics events. Status, error, duration, and token deltas also remain
analytics fields rather than request-header fields.
Future detached-memory attribution using a real initiating turn ID as
`trigger_turn_id` is intentionally not part of this PR.
## Sync With Main
- Final pushed head `716342e79` is rebased onto `origin/main@0d37db4b2`.
- The metadata conflict came from upstream `#24160`, which added
`forked_from_thread_id` on the same `turn_metadata` surface. Resolution
preserves that field and its protection from client metadata override
alongside this PR's request-kind, compaction, and window-id fields.
- While resolving the overlapping commits, I removed an accidental
recursive model-request overlay and a duplicate detached-memory header
builder before completing the rebase.
## Latency / User Experience Boundary
- Foreground turns perform no new filesystem, git, or network work. New
fields are inserted into metadata already serialized for outgoing
requests.
- Compaction issues the same model/HTTP requests with the same prompt,
model, service tier, and sampling settings; only metadata bytes change.
- Startup prewarm already sent metadata; it is now correctly classified
as `prewarm`.
- Non-git detached memory now sends a small kind-only metadata header
rather than no header.
- This client diff adds no user-visible latency mechanism beyond
negligible serialization and header bytes on already-existing requests.
## Validation
On conflict-resolved head `1d35c2cfb` based on `origin/main@487521733`:
- `just fmt` (passed)
- `just fix -p codex-core` (passed)
- `git diff --check origin/main...HEAD` (passed)
- `just test -p codex-core -E 'test(turn_metadata) |
test(websocket_first_turn_uses_startup_prewarm_and_create) |
test(responses_stream_includes_turn_metadata_header_for_git_workspace_e2e)
|
test(responses_websocket_forwards_turn_metadata_on_initial_and_incremental_create)
| test(remote_compact_v2_retries_failures_with_stream_retry_budget) |
test(window_id_advances_after_compact_persists_on_resume_and_resets_on_fork)'`
(`23 passed`; `bench-smoke` passed)
- `just test -p codex-app-server -E
'test(turn_start_forwards_client_metadata_to_responses_request_v2) |
test(turn_start_forwards_client_metadata_to_responses_websocket_request_body_v2)
| test(auto_compaction_remote_emits_started_and_completed_items)'` (`3
passed`; `bench-smoke` passed)
- `just test -p codex-memories-write` (`29 passed`; `bench-smoke`
passed)
## Why
The turn loop no longer needs to decide when a `ModelClientSession`
should reset its websocket state after compaction. That reset behavior
belongs inside the model client, where the websocket cache and retry
state are owned. The repo guidance now calls this out explicitly so
future changes let the incremental request logic decide whether the
previous request can be reused.
## What Changed
- Removed the `reset_client_session` return value from pre-sampling and
auto-compact helpers in `core/src/session/turn.rs`.
- Changed compaction helpers to return `CodexResult<()>` so callers only
handle success or failure.
- Made `ModelClientSession::reset_websocket_session` private to
`core/src/client.rs`, leaving it callable only from model-client
internals.
- Added `AGENTS.md` guidance not to call `reset_client_session`
unnecessarily.
## Validation
- `just test -p codex-core session::turn`
## Why
`prewarm_websocket` intentionally stays out of rollout inference
tracing, but the next traced websocket request can still reuse the
warmup `response_id` and send an empty `input` delta. If tracing records
that wire payload verbatim, replay sees an incremental request whose
parent was never traced and cannot reconstruct the conversation.
This fixes that at the producer boundary instead of relaxing
`rollout-trace` replay semantics around unresolved
`previous_response_id` values.
## What
- track whether the last websocket response came from an untraced warmup
and clear that state when the websocket session is reset or reconnected
- when a traced websocket request reuses that warmup parent, keep
sending the compressed websocket request on the wire but record the
logical `ResponsesApiRequest` in the rollout trace
- add a regression test that proves replay reconstructs the logical user
message even though the websocket follow-up carries
`previous_response_id = warm-1` with empty `input`
- update `InferenceTraceAttempt::record_started` docs to reflect that
callers may record a logical request rather than the exact transport
payload
## Testing
- `cargo test -p codex-core --test all
responses_websocket_request_prewarm_traces_logical_request`
## Why
Model catalog responses can now advertise a nullable
`default_service_tier` for each model. Codex needs to preserve three
distinct states all the way from config/app-server inputs to inference:
- no explicit service tier, so the client may apply the current model
catalog default when FastMode is enabled
- explicit `default`, meaning the user intentionally wants standard
routing
- explicit catalog tier ids such as `priority`, `flex`, or future tiers
Keeping those states distinct prevents the UI from showing one tier
while core sends another, especially after model switches or app-server
`thread/start` / `turn/start` updates.
## What Changed
- Plumbed `default_service_tier` through model catalog protocol types,
app-server model responses, generated schemas, model cache fixtures, and
provider/model-manager conversions.
- Added the request-only `default` service tier sentinel and normalized
legacy config spelling so `fast` in `config.toml` still materializes as
the runtime/request id `priority`.
- Moved catalog default resolution to the TUI/client side, including
recomputing the effective service tier when model/FastMode-dependent
surfaces change.
- Updated app-server thread lifecycle config construction so
`serviceTier: null` preserves explicit standard-routing intent by
mapping to `default` instead of internal `None`.
- Kept core responsible for validating explicit tiers against the
current model and stripping `default` before `/v1/responses`, without
applying catalog defaults itself.
## Validation
- `CARGO_INCREMENTAL=0 cargo build -p codex-cli`
- `CARGO_INCREMENTAL=0 cargo test -p codex-app-server model_list`
- `cargo test -p codex-tui service_tier`
- `cargo test -p codex-protocol service_tier_for_request`
- `cargo test -p codex-core get_service_tier`
- `RUST_MIN_STACK=8388608 CARGO_INCREMENTAL=0 cargo test -p codex-core
service_tier`
## Why
Remote compaction currently sends a unary `POST /responses/compact` and
waits for the full response before replacing history or emitting the
completed `ContextCompaction` item. Unlike normal `/responses` streaming
requests, this unary compact request had no timeout boundary. If the
backend accepts the request and then stalls before returning a body, the
existing request retry policy never sees a transport error, so the
compact turn can remain stuck after the started item with no completion
or actionable error.
That matches the reported hang shape in issues such as #18363, where
logs show `responses/compact` was posted but no corresponding compact
completion followed. A bounded request timeout gives the existing retry
policy a concrete timeout error to retry instead of letting the user sit
indefinitely on automatic context compaction.
## What
- Add a request timeout to legacy `/responses/compact` calls.
- Size that timeout from the provider stream idle timeout with a
conservative multiplier, so the default compact attempt gets 20 minutes
rather than the 5 minute stream idle window.
- Map API transport timeouts to a request timeout error instead of the
child-process timeout message.
## Testing
- Not run (per request; CI will cover).
## Why
`model_auto_compact_token_limit` has only been able to budget the full
active context. That makes it hard to set a small "growth since
compaction" budget for sessions that preserve a large carried window
prefix: the preserved prefix can consume the whole budget and force
immediate repeated compaction.
This PR adds an opt-in `body_after_prefix` scope so callers can apply
`model_auto_compact_token_limit` to sampled output and later growth
after the current carried prefix, while still forcing compaction before
the full model context window is exhausted.
## What changed
- Adds `AutoCompactTokenLimitScope` with the existing `total` behavior
as the default and a new `body_after_prefix` mode:
[`config_types.rs`](https://github.com/openai/codex/blob/973806b1cb35792555bead994cb3ed94656eb171/codex-rs/protocol/src/config_types.rs#L24-L37).
- Threads `model_auto_compact_token_limit_scope` through config loading,
`Config`, `core-api`, and app-server v2 schema/TypeScript generation.
- Records the first observed input-token count for a `body_after_prefix`
compaction window and uses it as the baseline when deciding whether the
scoped auto-compaction budget is exhausted:
[`turn.rs`](https://github.com/openai/codex/blob/973806b1cb35792555bead994cb3ed94656eb171/codex-rs/core/src/session/turn.rs#L743-L781).
- Keeps a hard context-window cap in `body_after_prefix`, so scoped
budgeting cannot let the active context overrun the usable window.
## Verification
Added compact-suite coverage for the two key behaviors:
`body_after_prefix` does not re-compact just because the carried prefix
is larger than the scoped budget, and it still compacts when the total
active context reaches the configured context window:
[`compact.rs`](https://github.com/openai/codex/blob/973806b1cb35792555bead994cb3ed94656eb171/codex-rs/core/tests/suite/compact.rs#L3003-L3128).
## Why
Compaction now installs replacement history inside the session, but the
turn and compaction callers were still reaching into
`ModelClientSession` to reset websocket transport state after that
install. That made a transport-level reset part of the compaction API
even though websocket incremental request selection already checks
whether the next request is a strict extension of the previous one and
falls back to a full `response.create` when it is not.
## What changed
- Removed the compaction-side calls to `reset_websocket_session` from
`compact.rs` and `session/turn.rs`.
- Simplified pre-sampling and mid-turn compaction helpers so they return
`CodexResult<()>` instead of carrying a reset flag.
- Made `ModelClientSession::reset_websocket_session` private to
`client.rs`, leaving only the websocket timeout recovery path inside the
client as a caller.
## Validation
- `cargo test -p codex-core --test all
responses_websocket_creates_on_non_prefix`
- `cargo test -p codex-core --test all
steered_user_input_waits_for_model_continuation_after_mid_turn_compact`
- `cargo test -p codex-core --test all
pre_sampling_compact_runs_on_switch_to_smaller_context_model`
## Why
`CODEX_RS_SSE_FIXTURE` let integration-style CLI, exec, and TUI tests
bypass the normal Responses transport by reading SSE from local files.
That kept test-only behavior wired through production client code. The
affected tests can stay hermetic by using the existing
`core_test_support::responses` mock server and passing `openai_base_url`
instead.
## What Changed
- Removed the `CODEX_RS_SSE_FIXTURE` flag,
`codex_api::stream_from_fixture`, the `env-flags` dependency, and the
checked-in SSE fixture files.
- Repointed the affected core, exec, and TUI tests at `MockServer` with
the existing SSE event constructors.
- Removed the Bazel test data plumbing for the deleted fixtures and
refreshed cargo/Bazel lock state.
## Verification
- `cargo build -p codex-cli`
- `cargo test -p codex-api`
- `cargo test -p codex-core --test all responses_api_stream_cli`
- `cargo test -p codex-core --test all
integration_creates_and_checks_session_file`
- `cargo test -p codex-exec --test all ephemeral`
- `cargo test -p codex-exec --test all resume`
- `cargo test -p codex-tui --test all
resume_startup_does_not_consume_model_availability_nux_count`
- `just bazel-lock-update`
- `just bazel-lock-check`
- `just fix -p codex-api -p codex-core -p codex-exec -p codex-tui`
- `git diff --check`
## Summary
TL;DR: teaches `codex-rs` / app-server to request a desktop-provided
attestation token and attach it as `x-oai-attestation` on the scoped
ChatGPT Codex request paths.
## Details
This PR teaches the Codex app-server runtime how to request and attach
an attestation token. It does not generate DeviceCheck tokens directly;
instead, it relies on the connected desktop app to advertise that it can
generate attestation and then asks that app for a fresh header value
when needed.
The flow is:
1. The Codex desktop app connects to app-server.
2. During `initialize`, the app can advertise that it supports
`requestAttestation`.
3. Before app-server calls selected ChatGPT Codex endpoints, it sends
the internal server request `attestation/generate` to the app.
4. app-server receives a pre-encoded header value back.
5. app-server forwards that value as `x-oai-attestation` on the scoped
outbound requests.
The code in this repo is mostly protocol and runtime plumbing: it adds
the app-server request/response shape, introduces an attestation
provider in core, wires that provider into Responses / compaction /
realtime setup paths, and covers the intended scoping with tests. The
signed macOS DeviceCheck generation remains owned by the desktop app PR.
## Related PR
- Codex desktop app implementation:
https://github.com/openai/openai/pull/878649
## Validation
<details>
<summary>Tests run</summary>
```sh
cargo test -p codex-app-server-protocol
cargo test -p codex-core attestation --lib
cargo test -p codex-app-server --lib attestation
```
Also ran:
```sh
just fix -p codex-core
just fix -p codex-app-server
just fix -p codex-app-server-protocol
just fmt
just write-app-server-schema
```
</details>
<details>
<summary>E2E DeviceCheck validation</summary>
First validated the signed desktop app boundary directly: launched a
packaged signed `Codex.app`, sent `attestation/generate`, decoded the
returned `v1.` attestation header, and validated the extracted
DeviceCheck token with `personal/jm/verify_devicecheck_token.py` using
bundle ID `com.openai.codex`. Apple returned `status_code: 200` and
`is_ok: true`.
Then ran the fuller app + app-server flow. The packaged `Codex.app`
launched a current-branch app-server via `CODEX_CLI_PATH`, and a local
MITM proxy intercepted outbound `chatgpt.com` traffic. The app-server
requested `attestation/generate` from the real Electron app process, and
the intercepted `/backend-api/codex/responses` traffic included
`x-oai-attestation` on both routes:
```text
GET /backend-api/codex/responses Upgrade: websocket x-oai-attestation: present
POST /backend-api/codex/responses Upgrade: none x-oai-attestation: present
```
The captured header decoded to a DeviceCheck token that also validated
with Apple for `com.openai.codex` (`status_code: 200`, `is_ok: true`,
team `2DC432GLL2`).
</details>
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
`/fast` was wired as a one-off slash command even though model metadata
now exposes service tiers as catalog data. That meant adding another
tier, such as a slower/cheaper tier, would require more hardcoded TUI
plumbing instead of letting the model catalog drive the available
commands.
This change makes service-tier commands data-driven: each advertised
`service_tiers` entry becomes a `/name` command using the catalog
description, while the request path sends the tier `id` only when the
selected model supports it.
## What Changed
- Removed the hardcoded `/fast` slash-command variant and introduced
dynamic service-tier command items in the composer and command popup.
- Added toggle behavior for service-tier commands: invoking `/name`
selects that tier, and invoking it again clears the selection.
- Preserved the existing Fast-mode keybinding/status affordances by
resolving the current model tier whose name is `fast`, while still
sending the tier request value such as `priority`.
- Persisted service-tier selections as raw request strings so non-fast
tiers can round-trip through config.
- Updated the Bedrock catalog entry to advertise fast support through
`service_tiers` with `id: "priority"` and `name: "fast"`.
- Added defensive filtering in core so unsupported selected service
tiers are omitted from `/responses` requests.
## Validation
- Added/updated coverage for dynamic service-tier slash command lookup,
popup descriptions, composer dispatch, TUI fast toggling, and
unsupported-tier omission in core request construction.
- Local tests were not run per request.
---------
Co-authored-by: Codex <noreply@openai.com>
## Summary
- Add a `response.processed` websocket request payload and sender for
Responses API websockets.
- Send `response.processed` from `try_run_sampling_request` after a
response completes, local turn processing succeeds, and the
session-owned feature flag is enabled.
- Add websocket coverage for both enabled and disabled feature-flag
behavior.
## Validation
- `just fmt`
- `cargo test -p codex-core response_processed`
- `cargo test -p codex-api responses_websocket`
- `cargo test -p codex-features
responses_websocket_response_processed_is_under_development`
- `git diff --check`
- `just fix -p codex-api -p codex-core -p codex-features`
- `git diff --check origin/main...HEAD`
## Why
`/responses/compact` should preserve the request-affinity fields that
apply to the active auth mode. ChatGPT-auth compact requests need the
effective `service_tier`, and compact requests for every auth mode need
the stable `prompt_cache_key`, so compaction does not quietly lose
routing or cache behavior that normal sampling already has.
This follows the request-parity direction from #20719, but keeps the net
change focused on the compact payload fields needed here.
## What changed
- Add `service_tier` and `prompt_cache_key` to the compact endpoint
input payload.
- Build the remote compact payload from the existing responses request
builder output so `Fast` still maps to `priority` when compact sends a
service tier.
- Pass the turn service tier into remote compaction, but only include it
in compact payloads for ChatGPT-backed auth.
- Keep `prompt_cache_key` on compact payloads for all auth modes.
- Add request-body diff snapshot coverage in
`core/tests/suite/compact_remote.rs` for:
- API-key auth reusing `prompt_cache_key` while omitting `service_tier`
even when `Fast` is configured.
- ChatGPT auth reusing both `service_tier` and `prompt_cache_key`.
- Drive the snapshot coverage through five varied turns: plain text,
multi-part text, tool-call continuation, image+text input, local-shell
continuation, and final-turn reasoning output.
## Verification
- Added insta snapshots for compact request-body parity against the last
normal `/responses` request after five varied turns.
- Not run locally per repo guidance; relying on GitHub CI for test
execution.
---------
Co-authored-by: Codex <noreply@openai.com>
## Summary
Related to
https://openai.slack.com/archives/C095U48JNL9/p1777537279707449
TLDR:
We update the meaning of session ids and thread ids:
* thread_id stays as now
* session_id become a shared id between every thread under a /root
thread (i.e. every sub-agent share the same session id)
This PR introduces an explicit `SessionId` and threads it through the
protocol/client boundary so `session_id` and `thread_id` can diverge
when they need to, while preserving compatibility for older serialized
`session_configured` events.
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
Feedback reports do not currently surface a direct pointer to the last
model call, so investigations may require searching through many
requests in a session to find the bad response. Preserve the last
model-side IDs at response-stream time so immediate feedback reports
carry that breadcrumb.
## What changed
- Record `last_model_request_id` when a Responses stream exposes an
upstream request ID.
- Record `last_model_response_id` when the model response completes.
- Add unit coverage for the emitted feedback tags.
## Verification
- `cargo test -p codex-core
client::tests::response_stream_records_last_model_feedback_ids`
## Why
`ModelClientSession` and `compact_conversation_history()` were still
rebuilding the same `ResponsesApiRequest` fields separately. That
duplication makes it easy for normal `/responses` turns and compact
requests to drift when request-shape changes land later, which is
exactly the kind of cache-affecting divergence we want to avoid.
This follow-up keeps the scope small by extracting the shared
request-construction logic into one helper and using it from both paths.
## What changed
- move `ResponsesApiRequest` construction into a shared
`ModelClient::build_responses_request(...)` helper in
`core/src/client.rs`
- update the normal `/responses` streaming path to call that helper
instead of the old `ModelClientSession`-local implementation
- update `compact_conversation_history()` to derive its compact payload
from the same helper so `model`, `instructions`, `input`, `tools`,
`parallel_tool_calls`, `reasoning`, and `text` stay aligned with normal
request building
- add a unit test covering the shared helper's prompt cache key,
installation metadata, and `service_tier` behavior
## Verification
- `cargo test -p codex-core
build_responses_request_sets_shared_cache_and_metadata_fields`
- `cargo test -p codex-core --test all
remote_compact_v2_reuses_context_compaction_for_followups`
## Docs
No docs update needed.
Michael Bolin
