- include the requested sub-agent model and reasoning effort in the
spawn begin event\n- render that metadata next to the spawned agent name
and role in the TUI transcript
---------
Co-authored-by: Codex <noreply@openai.com>
## Summary
- add ARC monitor support for MCP tool calls by serializing MCP approval
requests into the ARC action shape and sending the relevant
conversation/policy context to the `/api/codex/safety/arc` endpoint
- route ARC outcomes back into MCP approval flow so `ask-user` falls
back to a user prompt and `steer-model` blocks the tool call, with
guardian/ARC tests covering the new request shape
- update the TUI approval copy from “Approve Once” to “Allow” / “Allow
for this session” and refresh the related
snapshots
---------
Co-authored-by: Fouad Matin <fouad@openai.com>
Co-authored-by: Fouad Matin <169186268+fouad-openai@users.noreply.github.com>
There are some bug investigations that currently require us to ask users
for their user ID even though they've already uploaded logs and session
details via `/feedback`. This frustrates users and increases the time
for diagnosis.
This PR includes the ChatGPT user ID in the metadata uploaded for
`/feedback` (both the TUI and app-server).
- collect input/output transcript deltas into active handoff transcript
state
- attach and clear that transcript on each handoff, and regenerate
schema/tests
(Experimental)
This PR adds a first MVP for hooks, with SessionStart and Stop
The core design is:
- hooks live in a dedicated engine under codex-rs/hooks
- each hook type has its own event-specific file
- hook execution is synchronous and blocks normal turn progression while
running
- matching hooks run in parallel, then their results are aggregated into
a normalized HookRunSummary
On the AppServer side, hooks are exposed as operational metadata rather
than transcript-native items:
- new live notifications: hook/started, hook/completed
- persisted/replayed hook results live on Turn.hookRuns
- we intentionally did not add hook-specific ThreadItem variants
Hooks messages are not persisted, they remain ephemeral. The context
changes they add are (they get appended to the user's prompt)
#### What
###### Context + Problem
With the introduction of plugins, we now have one more type of
`$`-mentionable item in the TUI's popup menu on `$`. Apps, skills, and
plugins can all have the same user-facing name, and we attempt to
distinguish with a category tag suffix, like `[App]`. This has a few
problems:
- We decide to show tags by the text that will be inserted into the
conversation, not the actual user-visible text, so two visibly-identical
entries can have no clarifying category tag suffix
- The category tag is a suffix and commonly gets cut off by long
descriptions
- The skill category tag is currently only displayed on repo skills as
`[Repo]`, which is confusing to most users
- The plugin category tag is currently `[<marketplace-name>]`, which is
also confusing to most users
###### Solution
- **Always** show a **prefix** category tag that is `[Skill]`, `[App]`,
or `[Plugin]`. No conditional rendering or copy.
Before:
<img width="801" height="153" alt="image"
src="https://github.com/user-attachments/assets/448e06e7-2af8-4c14-9804-ed1ca17cf514"
/>
After:
<img width="800" height="118" alt="image"
src="https://github.com/user-attachments/assets/57895b41-06fe-4d92-887b-68704c5a15fd"
/>
I also feel this clarifies the results at-a-glance while you scroll:
https://github.com/user-attachments/assets/cbdd5840-53d9-4656-812c-6e816755e1fd
### Tests
Added + updated tests (including snapshots), tested locally
## Summary
request_permissions flows should support persisting results for the
session.
Open Question: Still deciding if we need within-turn approvals - this
adds complexity but I could see it being useful
## Testing
- [x] Updated unit tests
---------
Co-authored-by: Codex <noreply@openai.com>
## Summary
- align the guardian permission test with the actual sandbox policy it
widens and use a slightly larger Windows-only timeout budget
- expose the additional-permissions normalization helper to the guardian
test module
- replace the guardian popup snapshot assertion with targeted string
assertions
## Why this fixes the flake
This group was carrying two separate sources of drift. The guardian core
test widened derived sandbox policies without updating the source
sandbox policy, and it used a Windows command/timeout combination that
was too tight on slower runners. Separately, the TUI test was
snapshotting the full popup even though unrelated feature text changes
were the only thing moving. The new assertions keep coverage on the
guardian entry itself while removing unrelated snapshot churn.
## Summary
- remove the remaining model-visible guardian-specific `on-request`
prompt additions so enabling the feature does not change the main
approval-policy instructions
- neutralize user-facing guardian wording to talk about automatic
approval review / approval requests rather than a second reviewer or
only sandbox escalations
- tighten guardian retry-context handling so agent-authored
`justification` stays in the structured action JSON and is not also
injected as raw retry context
- simplify guardian review plumbing in core by deleting dead
prompt-append paths and trimming some request/transcript setup code
## Notable Changes
- delete the dead `permissions/approval_policy/guardian.md` append path
and stop threading `guardian_approval_enabled` through model-facing
developer-instruction builders
- rename the experimental feature copy to `Automatic approval review`
and update the `/experimental` snapshot text accordingly
- make approval-review status strings generic across shell, patch,
network, and MCP review types
- forward real sandbox/network retry reasons for shell and unified-exec
guardian review, but do not pass agent-authored justification as raw
retry context
- simplify `guardian.rs` by removing the one-field request wrapper,
deduping reasoning-effort selection, and cleaning up transcript entry
collection
## Testing
- `just fmt`
- full validation left to CI
---------
Co-authored-by: Codex <noreply@openai.com>
Adds a built-in `request_permissions` tool and wires it through the
Codex core, protocol, and app-server layers so a running turn can ask
the client for additional permissions instead of relying on a static
session policy.
The new flow emits a `RequestPermissions` event from core, tracks the
pending request by call ID, forwards it through app-server v2 as an
`item/permissions/requestApproval` request, and resumes the tool call
once the client returns an approved subset of the requested permission
profile.
## Summary
- split the pending input preview into labeled pending-steer and queued
follow-up sections
- explain that pending steers submit after the next tool call and that
Esc can interrupt and send them immediately
- treat Esc as an interrupt-plus-resubmit path when pending steers
exist, with updated TUI snapshots and tests
Queues and steers:
<img width="1038" height="263" alt="Screenshot 2026-03-07 at 10 17
17 PM"
src="https://github.com/user-attachments/assets/4ef433ef-27a3-4b7c-ad69-2046f6eb89e6"
/>
After pressing Esc:
<img width="1046" height="320" alt="Screenshot 2026-03-07 at 10 17
21 PM"
src="https://github.com/user-attachments/assets/0f4d89e0-b6b9-486a-9f04-b6021f169ba7"
/>
## Codex author
`codex resume 019cc6f4-2cca-7803-b717-8264526dbd97`
---------
Co-authored-by: Codex <noreply@openai.com>
The TUI was showing the raw configured `model_context_window` until the
first
`TokenCount` event arrived, even though core had already emitted the
effective
runtime window on `TurnStarted`. This made the footer, status-line
context
window, and `/status` output briefly inconsistent for models/configs
where the
effective window differs from the configured value, such as the
`gpt-5.4`
1,000,000-token override reported in #13623.
Update the TUI to cache `TurnStarted.model_context_window` immediately
so
pre-token-count displays use the runtime effective window, and add
regression
coverage for the startup path.
---------
Co-authored-by: Charles Cunningham <ccunningham@openai.com>
Co-authored-by: Codex <noreply@openai.com>
## Summary
#13910 was merged with some unused imports, let's fix this
## Testing
- [x] Let's make sure CI is green
---------
Co-authored-by: Charles Cunningham <ccunningham@openai.com>
Co-authored-by: Codex <noreply@openai.com>
## Summary
- add the guardian reviewer flow for `on-request` approvals in command,
patch, sandbox-retry, and managed-network approval paths
- keep guardian behind `features.guardian_approval` instead of exposing
a public `approval_policy = guardian` mode
- route ordinary `OnRequest` approvals to the guardian subagent when the
feature is enabled, without changing the public approval-mode surface
## Public model
- public approval modes stay unchanged
- guardian is enabled via `features.guardian_approval`
- when that feature is on, `approval_policy = on-request` keeps the same
approval boundaries but sends those approval requests to the guardian
reviewer instead of the user
- `/experimental` only persists the feature flag; it does not rewrite
`approval_policy`
- CLI and app-server no longer expose a separate `guardian` approval
mode in this PR
## Guardian reviewer
- the reviewer runs as a normal subagent and reuses the existing
subagent/thread machinery
- it is locked to a read-only sandbox and `approval_policy = never`
- it does not inherit user/project exec-policy rules
- it prefers `gpt-5.4` when the current provider exposes it, otherwise
falls back to the parent turn's active model
- it fail-closes on timeout, startup failure, malformed output, or any
other review error
- it currently auto-approves only when `risk_score < 80`
## Review context and policy
- guardian mirrors `OnRequest` approval semantics rather than
introducing a separate approval policy
- explicit `require_escalated` requests follow the same approval surface
as `OnRequest`; the difference is only who reviews them
- managed-network allowlist misses that enter the approval flow are also
reviewed by guardian
- the review prompt includes bounded recent transcript history plus
recent tool call/result evidence
- transcript entries and planned-action strings are truncated with
explicit `<guardian_truncated ... />` markers so large payloads stay
bounded
- apply-patch reviews include the full patch content (without
duplicating the structured `changes` payload)
- the guardian request layout is snapshot-tested using the same
model-visible Responses request formatter used elsewhere in core
## Guardian network behavior
- the guardian subagent inherits the parent session's managed-network
allowlist when one exists, so it can use the same approved network
surface while reviewing
- exact session-scoped network approvals are copied into the guardian
session with protocol/port scope preserved
- those copied approvals are now seeded before the guardian's first turn
is submitted, so inherited approvals are available during any immediate
review-time checks
## Out of scope / follow-ups
- the sandbox-permission validation split was pulled into a separate PR
and is not part of this diff
- a future follow-up can enable `serde_json` preserve-order in
`codex-core` and then simplify the guardian action rendering further
---------
Co-authored-by: Codex <noreply@openai.com>
Addresses feature request #13660
Adds new option to `/statusline` so the status line can display "fast
on" or "fast off"
Summary
- introduce a `FastMode` status-line item so `/statusline` can render
explicit `Fast on`/`Fast off` text for the service tier
- wire the item into the picker metadata and resolve its string from
`ChatWidget` without adding any unrelated `thread-name` logic or storage
changes
- ensure the refresh paths keep the cached footer in sync when the
service tier (fast mode) changes
Testing
- Manually tested
Here's what it looks like when enabled:
<img width="366" height="75" alt="image"
src="https://github.com/user-attachments/assets/7f992d2b-6dab-49ed-aa43-ad496f56f193"
/>
## Summary
- treat `requirements.toml` `allowed_domains` and `denied_domains` as
managed network baselines for the proxy
- in restricted modes by default, build the effective runtime policy
from the managed baseline plus user-configured allowlist and denylist
entries, so common hosts can be pre-approved without blocking later user
expansion
- add `experimental_network.managed_allowed_domains_only = true` to pin
the effective allowlist to managed entries, ignore user allowlist
additions, and hard-deny non-managed domains without prompting
- apply `managed_allowed_domains_only` anywhere managed network
enforcement is active, including full access, while continuing to
respect denied domains from all sources
- add regression coverage for merged-baseline behavior, managed-only
behavior, and full-access managed-only enforcement
## Behavior
Assuming `requirements.toml` defines both
`experimental_network.allowed_domains` and
`experimental_network.denied_domains`.
### Default mode
- By default, the effective allowlist is
`experimental_network.allowed_domains` plus user or persisted allowlist
additions.
- By default, the effective denylist is
`experimental_network.denied_domains` plus user or persisted denylist
additions.
- Allowlist misses can go through the network approval flow.
- Explicit denylist hits and local or private-network blocks are still
hard-denied.
- When `experimental_network.managed_allowed_domains_only = true`, only
managed `allowed_domains` are respected, user allowlist additions are
ignored, and non-managed domains are hard-denied without prompting.
- Denied domains continue to be respected from all sources.
### Full access
- With managed requirements present, the effective allowlist is pinned
to `experimental_network.allowed_domains`.
- With managed requirements present, the effective denylist is pinned to
`experimental_network.denied_domains`.
- There is no allowlist-miss approval path in full access.
- Explicit denylist hits are hard-denied.
- `experimental_network.managed_allowed_domains_only = true` now also
applies in full access, so managed-only behavior remains in effect
anywhere managed network enforcement is active.
## Summary
This is a purely mechanical refactor of `OtelManager` ->
`SessionTelemetry` to better convey what the struct is doing. No
behavior change.
## Why
`OtelManager` ended up sounding much broader than what this type
actually does. It doesn't manage OTEL globally; it's the session-scoped
telemetry surface for emitting log/trace events and recording metrics
with consistent session metadata (`app_version`, `model`, `slug`,
`originator`, etc.).
`SessionTelemetry` is a more accurate name, and updating the call sites
makes that boundary a lot easier to follow.
## Validation
- `just fmt`
- `cargo test -p codex-otel`
- `cargo test -p codex-core`
Enhance pty utils:
* Support closing stdin
* Separate stderr and stdout streams to allow consumers differentiate them
* Provide compatibility helper to merge both streams back into combined one
* Support specifying terminal size for pty, including on-demand resizes while process is already running
* Support terminating the process while still consuming its outputs
#### What
Add structured `@plugin` parsing and TUI support for plugin mentions.
- Core: switch from plain-text `@display_name` parsing to structured
`plugin://...` mentions via `UserInput::Mention` and
`[$...](plugin://...)` links in text, same pattern as apps/skills.
- TUI: add plugin mention popup, autocomplete, and chips when typing
`$`. Load plugin capability summaries and feed them into the composer;
plugin mentions appear alongside skills and apps.
- Generalize mention parsing to a sigil parameter, still defaults to `$`
<img width="797" height="119" alt="image"
src="https://github.com/user-attachments/assets/f0fe2658-d908-4927-9139-73f850805ceb"
/>
Builds on #13510. Currently clients have to build their own `id` via
`plugin@marketplace` and filter plugins to show by `enabled`, but we
will add `id` and `available` as fields returned from `plugin/list`
soon.
####Tests
Added tests, verified locally.
### Summary
This adds turn-level latency metrics for the first model output and the
first completed agent message.
- `codex.turn.ttft.duration_ms` starts at turn start and records on the
first output signal we see from the model. That includes normal
assistant text, reasoning deltas, and non-text outputs like tool-call
items.
- `codex.turn.ttfm.duration_ms` also starts at turn start, but it
records when the first agent message finishes streaming rather than when
its first delta arrives.
### Implementation notes
The timing is tracked in codex-core, not app-server, so the definition
stays consistent across CLI, TUI, and app-server clients.
I reused the existing turn lifecycle boundary that already drives
`codex.turn.e2e_duration_ms`, stored the turn start timestamp in turn
state, and record each metric once per turn.
I also wired the new metric names into the OTEL runtime metrics summary
so they show up in the same in-memory/debug snapshot path as the
existing timing metrics.
## Summary
Simplify the trusted directory flow. This logic was originally designed
several months ago, to determine if codex should start in read-only or
workspace-write mode. However, that's no longer the purpose of directory
trust - and therefore we should get rid of this logic.
## Testing
- [x] Unit tests pass
## Summary
- delete the network proxy admin server and its runtime listener/task
plumbing
- remove the admin endpoint config, runtime, requirement, protocol,
schema, and debug-surface fields
- update proxy docs to reflect the remaining HTTP and SOCKS listeners
only
## Summary
- default the resume picker sort key to UpdatedAt instead of CreatedAt
- keep Tab sort toggling behavior and update the test expectation for
the new default
## Testing
- just fmt
- cargo test -p codex-tui
Co-authored-by: Codex <noreply@openai.com>
## Summary
- render pending steer previews with a single `pending steer:` prefix
instead of repeating it for each source line
- reuse the same truncation path for pending steers and queued drafts so
multiline previews behave consistently
- add snapshot coverage for the multiline pending steer case
Before
<img width="969" height="219" alt="Screenshot 2026-03-05 at 3 55 11 PM"
src="https://github.com/user-attachments/assets/b062c9c8-43d3-4a52-98e0-3c7643d1697b"
/>
After
<img width="965" height="203" alt="Screenshot 2026-03-05 at 3 56 08 PM"
src="https://github.com/user-attachments/assets/40935863-55b3-444f-9e14-1ac63126b2e1"
/>
## Codex author
`codex resume 019cc054-385e-79a3-bb85-ec9499623bd8`
Co-authored-by: Codex <noreply@openai.com>
- Update `models.json` to surface the new model entry.
- Refresh the TUI model picker snapshot to match the updated catalog
ordering.
---------
Co-authored-by: aibrahim-oai <219906144+aibrahim-oai@users.noreply.github.com>
## Note-- added plugin mentions via @, but that conflicts with file
mentions
depends and builds upon #13433.
- introduces explicit `@plugin` mentions. this injects the plugin's mcp
servers, app names, and skill name format into turn context as a dev
message.
- we do not yet have UI for these mentions, so we currently parse raw
text (as opposed to skills and apps which have UI chips, autocomplete,
etc.) this depends on a `plugins/list` app-server endpoint we can feed
the UI with, which is upcoming
- also annotate mcp and app tool descriptions with the plugin(s) they
come from. this gives the model a first class way of understanding what
tools come from which plugins, which will help implicit invocation.
### Tests
Added and updated tests, unit and integration. Also confirmed locally a
raw `@plugin` injects the dev message, and the model knows about its
apps, mcps, and skills.
This PR adds a durable trace linkage for each turn by storing the active
trace ID on the rollout TurnContext record stored in session rollout
files.
Before this change, we propagated trace context at runtime but didn’t
persist a stable per-turn trace key in rollout history. That made
after-the-fact debugging harder (for example, mapping a historical turn
to the corresponding trace in datadog). This sets us up for much easier
debugging in the future.
### What changed
- Added an optional `trace_id` to TurnContextItem (rollout schema).
- Added a small OTEL helper to read the current span trace ID.
- Captured `trace_id` when creating `TurnContext` and included it in
`to_turn_context_item()`.
- Updated tests and fixtures that construct TurnContextItem so
older/no-trace cases still work.
### Why this approach
TurnContext is already the canonical durable per-turn metadata in
rollout. This keeps ownership clean: trace linkage lives with other
persisted turn metadata.
This adds a first-class server request for MCP server elicitations:
`mcpServer/elicitation/request`.
Until now, MCP elicitation requests only showed up as a raw
`codex/event/elicitation_request` event from core. That made it hard for
v2 clients to handle elicitations using the same request/response flow
as other server-driven interactions (like shell and `apply_patch`
tools).
This also updates the underlying MCP elicitation request handling in
core to pass through the full MCP request (including URL and form data)
so we can expose it properly in app-server.
### Why not `item/mcpToolCall/elicitationRequest`?
This is because MCP elicitations are related to MCP servers first, and
only optionally to a specific MCP tool call.
In the MCP protocol, elicitation is a server-to-client capability: the
server sends `elicitation/create`, and the client replies with an
elicitation result. RMCP models it that way as well.
In practice an elicitation is often triggered by an MCP tool call, but
not always.
### What changed
- add `mcpServer/elicitation/request` to the v2 app-server API
- translate core `codex/event/elicitation_request` events into the new
v2 server request
- map client responses back into `Op::ResolveElicitation` so the MCP
server can continue
- update app-server docs and generated protocol schema
- add an end-to-end app-server test that covers the full round trip
through a real RMCP elicitation flow
- The new test exercises a realistic case where an MCP tool call
triggers an elicitation, the app-server emits
mcpServer/elicitation/request, the client accepts it, and the tool call
resumes and completes successfully.
### app-server API flow
- Client starts a thread with `thread/start`.
- Client starts a turn with `turn/start`.
- App-server sends `item/started` for the `mcpToolCall`.
- While that tool call is in progress, app-server sends
`mcpServer/elicitation/request`.
- Client responds to that request with `{ action: "accept" | "decline" |
"cancel" }`.
- App-server sends `serverRequest/resolved`.
- App-server sends `item/completed` for the mcpToolCall.
- App-server sends `turn/completed`.
- If the turn is interrupted while the elicitation is pending,
app-server still sends `serverRequest/resolved` before the turn
finishes.
- add a speed row to the startup/session header under the model row
- render the speed row with the same styling pattern as the model row,
using /fast to change
- show only Fast or Standard to users and update the affected snapshots
---------
Co-authored-by: Codex <noreply@openai.com>
