Summary
- detect skill-invoking shell commands based on the original command
string, request approvals when needed, and cache positive decisions per
session
- keep implicit skill invocation emitted after approval and keep skill
approval decline messaging centralized to the shell handler
- expand and adjust skill approval tests to cover shell-based skill
scripts while matching the new detection expectations
Testing
- Not run (not requested)
#### What
Try matching `\w+`-namespaced model after `longest prefix` as heuristic
to match `ModelInfo` from list of candidates.
This shouldn't regress existing behavior:
- `gpt-5.2-codex` -> `gpt-5.2` if `gpt-5.2-codex` not present
- `gpt-5.3` -> `gpt-5` if `gpt-5.3` not present
- `gpt-9` still doesn't match anything
while being more forgiving for custom prefixes:
- `oai/gpt-5.3-codex` -> `gpt-5.3-codex`
#### Tests
Added unit test.
## Summary
Introduces the initial implementation of Feature::RequestPermissions.
RequestPermissions allows the model to request that a command be run
inside the sandbox, with additional permissions, like writing to a
specific folder. Eventually this will include other rules as well, and
the ability to persist these permissions, but this PR is already quite
large - let's get the core flow working and go from there!
<img width="1279" height="541" alt="Screenshot 2026-02-15 at 2 26 22 PM"
src="https://github.com/user-attachments/assets/0ee3ec0f-02ec-4509-91a2-809ac80be368"
/>
## Testing
- [x] Added tests
- [x] Tested locally
- [x] Feature
Send a request with `generate: falls` but a full set of tools and
instructions to pre-warm inference.
---------
Co-authored-by: Codex <noreply@openai.com>
rm `PRESETS` list harcoded in `model_presets` as we now have bundled
`models.json` with equivalent info.
update logic to rely on bundled models instead, update tests.
## Summary
Persist network approval allow/deny decisions as `network_rule(...)`
entries in execpolicy (not proxy config)
It adds `network_rule` parsing + append support in `codex-execpolicy`,
including `decision="prompt"` (parse-only; not compiled into proxy
allow/deny lists)
- compile execpolicy network rules into proxy allow/deny lists and
update the live proxy state on approval
- preserve requirements execpolicy `network_rule(...)` entries when
merging with file-based execpolicy
- reject broad wildcard hosts (for example `*`) for persisted
`network_rule(...)`
## Why
The zsh integration tests were still brittle in two ways:
- they relied on `CODEX_TEST_ZSH_PATH` / environment-specific setup, so
they often did not exercise the patched zsh fork that `shell-tool-mcp`
ships
- once the tests consistently used the vendored zsh fork, they exposed
real Linux-specific zsh-fork issues in CI
In particular, the Linux failures were not just test noise:
- the zsh-fork launch path was dropping `ExecRequest.arg0`, so Linux
`codex-linux-sandbox` arg0 dispatch did not run and zsh wrapper-mode
could receive malformed arguments
- the
`turn_start_shell_zsh_fork_subcommand_decline_marks_parent_declined_v2`
test uses the zsh exec bridge (which talks to the parent over a Unix
socket), but Linux restricted sandbox seccomp denies `connect(2)`,
causing timeouts on `ubuntu-24.04` x86/arm
This PR makes the zsh tests consistently run against the intended
vendored zsh fork and fixes/hardens the zsh-fork path so the Linux CI
signal is meaningful.
## What Changed
- Added a single shared test-only DotSlash file for the patched zsh fork
at `codex-rs/exec-server/tests/suite/zsh` (analogous to the existing
`bash` test resource).
- Updated both app-server and exec-server zsh tests to use that shared
DotSlash zsh (no duplicate zsh DotSlash file, no `CODEX_TEST_ZSH_PATH`
dependency).
- Updated the app-server zsh-fork test helper to resolve the shared
DotSlash zsh and avoid silently falling back to host zsh.
- Kept the app-server zsh-fork tests configured via `config.toml`, using
a test wrapper path where needed to force `zsh -df` (and rewrite `-lc`
to `-c`) for the subcommand-decline test.
- Hardened the app-server subcommand-decline zsh-fork test for CI
variability:
- tolerate an extra `/responses` POST with a no-op mock response
- tolerate non-target approval ordering while remaining strict on the
two `/usr/bin/true` approvals and decline behavior
- use `DangerFullAccess` on Linux for this one test because it validates
zsh approval flow, not Linux sandbox socket restrictions
- Fixed zsh-fork process launching on Linux by preserving `req.arg0` in
`ZshExecBridge::execute_shell_request(...)` so `codex-linux-sandbox`
arg0 dispatch continues to work.
- Moved `maybe_run_zsh_exec_wrapper_mode()` under
`arg0_dispatch_or_else(...)` in `app-server` and `cli` so wrapper-mode
handling coexists correctly with arg0-dispatched helper modes.
- Consolidated duplicated `dotslash -- fetch` resolution logic into
shared test support (`core/tests/common/lib.rs`).
- Updated `codex-rs/exec-server/tests/suite/accept_elicitation.rs` to
use DotSlash zsh and hardened the zsh elicitation test for Bazel/zsh
differences by:
- resolving an absolute `git` path
- running `git init --quiet .`
- asserting success / `.git` creation instead of relying on banner text
## Verification
- `cargo test -p codex-app-server turn_start_zsh_fork -- --nocapture`
- `cargo test -p codex-exec-server accept_elicitation -- --nocapture`
- `bazel test //codex-rs/exec-server:exec-server-all-test
--test_output=streamed --test_arg=--nocapture
--test_arg=accept_elicitation_for_prompt_rule_with_zsh`
- CI (`rust-ci`) on the final cleaned commit: `Tests — ubuntu-24.04 -
x86_64-unknown-linux-gnu` and `Tests — ubuntu-24.04-arm -
aarch64-unknown-linux-gnu` passed in [run
22291424358](https://github.com/openai/codex/actions/runs/22291424358)
## Why
`codex-core::all` has a flaky test,
`suite::realtime_conversation::conversation_start_audio_text_close_round_trip`,
that assumes a fixed ordering between `conversation.item.create` and
`response.input_audio.delta` requests.
That ordering is not guaranteed: realtime text and audio input are
forwarded through separate queues and a background task, so either
request can be observed first while still being correct behavior.
## What Changed
- Updated the assertion in
`codex-rs/core/tests/suite/realtime_conversation.rs` to compare the two
observed request types order-independently.
- Kept the existing checks that `session.create` is sent first and that
exactly two follow-up requests are recorded.
## Verification
- Re-ran `cargo test -p codex-core --test all
conversation_start_audio_text_close_round_trip` 10 times locally.
## Why
`codex-rs/core/src/lib.rs` re-exported a broad set of types and modules
from `codex-protocol` and `codex-shell-command`. That made it easy for
workspace crates to import those APIs through `codex-core`, which in
turn hides dependency edges and makes it harder to reduce compile-time
coupling over time.
This change removes those public re-exports so call sites must import
from the source crates directly. Even when a crate still depends on
`codex-core` today, this makes dependency boundaries explicit and
unblocks future work to drop `codex-core` dependencies where possible.
## What Changed
- Removed public re-exports from `codex-rs/core/src/lib.rs` for:
- `codex_protocol::protocol` and related protocol/model types (including
`InitialHistory`)
- `codex_protocol::config_types` (`protocol_config_types`)
- `codex_shell_command::{bash, is_dangerous_command, is_safe_command,
parse_command, powershell}`
- Migrated workspace Rust call sites to import directly from:
- `codex_protocol::protocol`
- `codex_protocol::config_types`
- `codex_protocol::models`
- `codex_shell_command`
- Added explicit `Cargo.toml` dependencies (`codex-protocol` /
`codex-shell-command`) in crates that now import those crates directly.
- Kept `codex-core` internal modules compiling by using `pub(crate)`
aliases in `core/src/lib.rs` (internal-only, not part of the public
API).
- Updated the two utility crates that can already drop a `codex-core`
dependency edge entirely:
- `codex-utils-approval-presets`
- `codex-utils-cli`
## Verification
- `cargo test -p codex-utils-approval-presets`
- `cargo test -p codex-utils-cli`
- `cargo check --workspace --all-targets`
- `just clippy`
## Why
Compiling `codex-rs/core` is a bottleneck for local iteration, so this
change continues the ongoing extraction of config-related functionality
out of `codex-core` and into `codex-config`.
The goal is not just to move code, but to reduce `codex-core` ownership
and indirection so more code depends on `codex-config` directly.
## What Changed
- Moved config diagnostics logic from
`core/src/config_loader/diagnostics.rs` into
`config/src/diagnostics.rs`.
- Updated `codex-core` to use `codex-config` diagnostics types/functions
directly where possible.
- Removed the `core/src/config_loader/diagnostics.rs` shim module
entirely; the remaining `ConfigToml`-specific calls are in
`core/src/config_loader/mod.rs`.
- Moved `CONFIG_TOML_FILE` into `codex-config` and updated existing
references to use `codex_config::CONFIG_TOML_FILE` directly.
- Added a direct `codex-config` dependency to `codex-cli` for its
`CONFIG_TOML_FILE` use.
## Summary
- move regular-turn context diff/full-context persistence into
`run_turn` so pre-turn compaction runs before incoming context updates
are recorded
- after successful pre-turn compaction, rely on a cleared
`reference_context_item` to trigger full context reinjection on the
follow-up regular turn (manual `/compact` keeps replacement history
summary-only and also clears the baseline)
- preserve `<model_switch>` when full context is reinjected, and inject
it *before* the rest of the full-context items
- scope `reference_context_item` and `previous_model` to regular user
turns only so standalone tasks (`/compact`, shell, review, undo) cannot
suppress future reinjection or `<model_switch>` behavior
- make context-diff persistence + `reference_context_item` updates
explicit in the regular-turn path, with clearer docs/comments around the
invariant
- stop persisting local `/compact` `RolloutItem::TurnContext` snapshots
(only regular turns persist `TurnContextItem` now)
- simplify resume/fork previous-model/reference-baseline hydration by
looking up the last surviving turn context from rollout lifecycle
events, including rollback and compaction-crossing handling
- remove the legacy fallback that guessed from bare `TurnContext`
rollouts without lifecycle events
- update compaction/remote-compaction/model-visible snapshots and
compact test assertions (including remote compaction mock response
shape)
## Why
We were persisting incoming context items before spawning the regular
turn task, which let pre-turn compaction requests accidentally include
incoming context diffs without the new user message. Fixing that exposed
follow-on baseline issues around `/compact`, resume/fork, and standalone
tasks that could cause duplicate context injection or suppress
`<model_switch>` instructions.
This PR re-centers the invariants around regular turns:
- regular turns persist model-visible context diffs/full reinjection and
update the `reference_context_item`
- standalone tasks do not advance those regular-turn baselines
- compaction clears the baseline when replacement history may have
stripped the referenced context diffs
## Follow-ups (TODOs left in code)
- `TODO(ccunningham)`: fix rollback/backtracking baseline handling more
comprehensively
- `TODO(ccunningham)`: include pending incoming context items in
pre-turn compaction threshold estimation
- `TODO(ccunningham)`: inject updated personality spec alongside
`<model_switch>` so some model-switch paths can avoid forced full
reinjection
- `TODO(ccunningham)`: review task turn lifecycle
(`TurnStarted`/`TurnComplete`) behavior and emit task-start context
diffs for task types that should have them (excluding `/compact`)
## Validation
- `just fmt`
- CI should cover the updated compaction/resume/model-visible snapshot
expectations and rollout-hydration behavior
- I did **not** rerun the full local test suite after the latest
resume-lookup / rollout-persistence simplifications
## Summary
`gpt-5.3-codex` really likes to write complicated shell scripts, and
suggest a partial prefix_rule that wouldn't actually approve the
command. We should only show the `prefix_rule` suggestion from the model
if it would actually fully approve the command the user is seeing.
This will technically cause more instances of overly-specific
suggestions when we fallback, but I think the UX is clearer,
particularly when the model doesn't necessarily understand the current
limitations of execpolicy parsing.
## Testing
- [x] Add unit tests
- [x] Add integration tests
- add top-level `experimental_realtime_ws_backend_prompt` config key
(experimental / do not use) and include it in config schema
- apply the override only to `Op::RealtimeConversation` websocket
`backend_prompt`, with config + realtime tests
- add top-level `experimental_realtime_ws_base_url` config key
(experimental / do not use) and include it in config schema
- apply the override only to `Op::RealtimeConversation` websocket
transport, with config + realtime tests
- Introduce `RealtimeConversationManager` for realtime API management
- Add `op::conversation` to start conversation, insert audio, insert
text, and close conversation.
- emit conversation lifecycle and realtime events.
- Move shared realtime payload types into codex-protocol and add core
e2e websocket tests for start/replace/transport-close paths.
Things to consider:
- Should we use the same `op::` and `Events` channel to carry audio? I
think we should try this simple approach and later we can create
separate one if the channels got congested.
- Sending text updates to the client: we can start simple and later
restrict that.
- Provider auth isn't wired for now intentionally
## Summary
- switch the review test SSE mock helper to use the shared hermetic mock
server setup
- ensure review tests always have a default `/v1/models` stub during
Codex session bootstrap
- remove the race that caused intermittent `/v1/models` connection
failures and flaky ETag refresh assertions
## Testing
- `just fmt`
- `cargo test -p codex-core --test all
refresh_models_on_models_etag_mismatch_and_avoid_duplicate_models_fetch`
- `cargo test -p codex-core --test all
review_uses_custom_review_model_from_config`
- repeated both targeted tests 5x in a loop
- `cargo clippy -p codex-core --tests -- -D warnings`
## Summary
Simplify network approvals by removing per-attempt proxy correlation and
moving to session-level approval dedupe keyed by (host, protocol, port).
Instead of encoding attempt IDs into proxy credentials/URLs, we now
treat approvals as a destination policy decision.
- Concurrent calls to the same destination share one approval prompt.
- Different destinations (or same host on different ports) get separate
prompts.
- Allow once approves the current queued request group only.
- Allow for session caches that (host, protocol, port) and auto-allows
future matching requests.
- Never policy continues to deny without prompting.
Example:
- 3 calls:
- a.com (line 443)
- b.com (line 443)
- a.com (line 443)
=> 2 prompts total (a, b), second a waits on the first decision.
- a.com:80 is treated separately from a.com line 443
## Testing
- `just fmt` (in `codex-rs`)
- `cargo test -p codex-core tools::network_approval::tests`
- `cargo test -p codex-core` (unit tests pass; existing
integration-suite failures remain in this environment)
Summary
- capture the origin for each configured MCP server and expose it via
the connection manager
- plumb MCP server name/origin into tool logging and emit
codex.tool_result events with those fields
- add unit coverage for origin parsing and extend OTEL tests to assert
empty MCP fields for non-MCP tools
- currently not logging full urls or url paths to prevent logging
potentially sensitive data
Testing
- Not run (not requested)
## Summary
Implements a configurable MCP OAuth callback URL override for `codex mcp
login` and app-server OAuth login flows, including support for non-local
callback endpoints (for example, devbox ingress URLs).
## What changed
- Added new config key: `mcp_oauth_callback_url` in
`~/.codex/config.toml`.
- OAuth authorization now uses `mcp_oauth_callback_url` as
`redirect_uri` when set.
- Callback handling validates the callback path against the configured
redirect URI path.
- Listener bind behavior is now host-aware:
- local callback URL hosts (`localhost`, `127.0.0.1`, `::1`) bind to
`127.0.0.1`
- non-local callback URL hosts bind to `0.0.0.0`
- `mcp_oauth_callback_port` remains supported and is used for the
listener port.
- Wired through:
- CLI MCP login flow
- App-server MCP OAuth login flow
- Skill dependency OAuth login flow
- Updated config schema and config tests.
## Why
Some environments need OAuth callbacks to land on a specific reachable
URL (for example ingress in remote devboxes), not loopback. This change
allows that while preserving local defaults for existing users.
## Backward compatibility
- No behavior change when `mcp_oauth_callback_url` is unset.
- Existing `mcp_oauth_callback_port` behavior remains intact.
- Local callback flows continue binding to loopback by default.
## Testing
- `cargo test -p codex-rmcp-client callback -- --nocapture`
- `cargo test -p codex-core --lib mcp_oauth_callback -- --nocapture`
- `cargo check -p codex-cli -p codex-app-server -p codex-rmcp-client`
## Example config
```toml
mcp_oauth_callback_port = 5555
mcp_oauth_callback_url = "https://<devbox>-<namespace>.gateway.<cluster>.internal.api.openai.org/callback"
TL;DR
Add top-level `model_catalog_json` config support so users can supply a
local model catalog override from a JSON file path (including adding new
models) without backend changes.
### Problem
Codex previously had no clean client-side way to replace/overlay model
catalog data for local testing of model metadata and new model entries.
### Fix
- Add top-level `model_catalog_json` config field (JSON file path).
- Apply catalog entries when resolving `ModelInfo`:
1. Base resolved model metadata (remote/fallback)
2. Catalog overlay from `model_catalog_json`
3. Existing global top-level overrides (`model_context_window`,
`model_supports_reasoning_summaries`, etc.)
### Note
Will revisit per-field overrides in a follow-up
### Tests
Added tests
Summary
- replace the stale `docs/config.md#feature-flags` reference in the
legacy feature notice with the canonical published URL
- align the deprecation notice test to expect the new link
This addresses #12123
We've continued to receive reports from users that they're seeing the
error message "Your access token could not be refreshed because your
refresh token was already used. Please log out and sign in again." This
PR fixes two holes in the token refresh logic that lead to this
condition.
Background: A previous change in token refresh introduced the
`UnauthorizedRecovery` object. It implements a state machine in the core
agent loop that first performs a load of the on-disk auth information
guarded by a check for matching account ID. If it finds that the on-disk
version has been updated by another instance of codex, it uses the
reloaded auth tokens. If the on-disk version hasn't been updated, it
issues a refresh request from the token authority.
There are two problems that this PR addresses:
Problem 1: We weren't doing the same thing for the code path used by the
app server interface. This PR effectively replicates the
`UnauthorizedRecovery` logic for that code path.
Problem 2: The `UnauthorizedRecovery` logic contained a hole in the
`ReloadOutcome::Skipped` case. Here's the scenario. A user starts two
instances of the CLI. Instance 1 is active (working on a task), instance
2 is idle. Both instances have the same in-memory cached tokens. The
user then runs `codex logout` or `codex login` to log in to a separate
account, which overwrites the `auth.json` file. Instance 1 receives a
401 and refreshes its token, but it doesn't write the new token to the
`auth.json` file because the account ID doesn't match. Instance 2 is
later activated and presented with a new task. It immediately hits a 401
and attempts to refresh its token but fails because its cached refresh
token is now invalid. To avoid this situation, I've changed the logic to
immediately fail a token refresh if the user has since logged out or
logged in to another account. This will still be seen as an error by the
user, but the cause will be clearer.
I also took this opportunity to clean up the names of existing functions
to make their roles clearer.
* `try_refresh_token` is renamed `request_chatgpt_token_refresh`
* the existing `refresh_token` is renamed `refresh_token_from_authority`
(there's a new higher-level function named `refresh_token` now)
* `refresh_tokens` is renamed `refresh_and_persist_chatgpt_token`, and
it now implicitly reloads
* `update_tokens` is renamed `persist_tokens`
## Summary
- add a dedicated `core/tests/suite/model_visible_layout.rs` snapshot
suite to materialize model-visible request layout in high-value
scenarios
- add three reviewer-focused snapshot scenarios:
- turn-level context updates (cwd / permissions / personality)
- first post-resume turn with model hydration + personality change
- first post-resume turn where pre-turn model override matches rollout
model
- wire the new suite into `core/tests/suite/mod.rs`
- commit generated `insta` snapshots under `core/tests/suite/snapshots/`
## Why
This creates a stable, reviewable baseline of model-visible context
layout against `main` before follow-on context-management refactors. It
lets subsequent PRs show focused snapshot diffs for behavior changes
instead of introducing the test surface and behavior changes at once.
## Testing
- `just fmt`
- `INSTA_UPDATE=always cargo test -p codex-core model_visible_layout`
## Summary
- standardize remote compaction test mocking around one default behavior
in shared helpers
- make default remote compact mocks mirror production shape: keep
`message/user` + `message/developer`, drop assistant/tool artifacts,
then append a summary user message
- switch non-special `compact_remote` tests to the shared default mock
instead of ad-hoc JSON payloads
## Special-case tests that still use explicit mocks
- remote compaction error payload / HTTP failure behavior
- summary-only compact output behavior
- manual `/compact` with no prior user messages
- stale developer-instruction injection coverage
## Why
This removes inconsistent manual remote compaction fixtures and gives us
one source of truth for normal remote compact behavior, while preserving
explicit mocks only where tests intentionally cover non-default
behavior.
zsh fork PR stack:
- https://github.com/openai/codex/pull/12051👈
- https://github.com/openai/codex/pull/12052
With upcoming support for a fork of zsh that allows us to intercept
`execve` and run execpolicy checks for each subcommand as part of a
`CommandExecution`, it will be possible for there to be multiple
approval requests for a shell command like `/path/to/zsh -lc 'git status
&& rg \"TODO\" src && make test'`.
To support that, this PR introduces a new `approval_id` field across
core, protocol, and app-server so that we can associate approvals
properly for subcommands.
### Summary
Ensure that we use the model value from the response header only so that
we are guaranteed with the correct slug name. We are no longer checking
against the model value from response so that we are less likely to have
false positive.
There are two different treatments - for SSE we use the header from the
response and for websocket we check top-level events.
rm `remote_models` feature flag.
We see issues like #11527 when a user has `remote_models` disabled, as
we always use the default fallback `ModelInfo`. This causes issues with
model performance.
Builds on #11690, which helps by warning the user when they are using
the default fallback. This PR will make that happen much less frequently
as an accidental consequence of disabling `remote_models`.
### Summary
Builiding off
5c75aa7b89 (diff-058ae8f109a8b84b4b79bbfa45f522c2233b9d9e139696044ae374d50b6196e0),
we have created a `model/rerouted` notification that captures the event
so that consumers can render as expected. Keep the `EventMsg::Warning`
path in core so that this does not affect TUI rendering.
`model/rerouted` is meant to be generic to account for future usage
including capacity planning etc.
## Summary
Fixes a few things in our exec_policy handling of prefix_rules:
1. Correctly match redirects specifically for exec_policy parsing. i.e.
if you have `prefix_rule(["echo"], decision="allow")` then `echo hello >
output.txt` should match - this should fix#10321
2. If there already exists any rule that would match our prefix rule
(not just a prompt), then drop it, since it won't do anything.
## Testing
- [x] Updated unit tests, added approvals ScenarioSpecs
Add per-turn notice when a request is downgraded to a fallback model due
to cyber safety checks.
**Changes**
- codex-api: Emit a ServerModel event based on the openai-model response
header and/or response payload (SSE + WebSocket), including when the
model changes mid-stream.
- core: When the server-reported model differs from the requested model,
emit a single per-turn warning explaining the reroute to gpt-5.2 and
directing users to Trusted
Access verification and the cyber safety explainer.
- app-server (v2): Surface these cyber model-routing warnings as
synthetic userMessage items with text prefixed by Warning: (and document
this behavior).
## Summary
This feature is now reasonably stable, let's remove it so we can
simplify our upcoming iterations here.
## Testing
- [x] Existing tests pass
