The Bazelification work in-flight over at
https://github.com/openai/codex/pull/8832 needs this fix so that Bazel
can find the path to the DotSlash file for `bash`.
With this change, the following almost works:
```
bazel test --test_output=errors //codex-rs/exec-server:exec-server-all-test
```
That is, now the `list_tools` test passes, but
`accept_elicitation_for_prompt_rule` still fails because it runs
Seatbelt itself, so it needs to be run outside Bazel's local sandboxing.
Skills discovery now follows symlink entries for SkillScope::User
($CODEX_HOME/skills) and SkillScope::Admin (e.g. /etc/codex/skills).
Added cycle protection: directories are canonicalized and tracked in a
visited set to prevent infinite traversal from circular links.
Added per-root traversal limits to avoid accidentally scanning huge
trees:
- max depth: 6
- max directories: 2000 (logs a warning if truncated)
For now, symlink stat failures and traversal truncation are logged
rather than surfaced as UI “invalid SKILL.md” warnings.
<img width="763" height="349" alt="Screenshot 2026-01-07 at 18 37 59"
src="https://github.com/user-attachments/assets/569d01cb-ea91-4113-889b-ba74df24adaf"
/>
It may not make sense to use the `/model` menu with a custom
OPENAI_BASE_URL. But some model proxies may support it, so we shouldn't
disable it completely. A warning is a reasonable compromise.
Fixes#8609
# Summary
Emphasize single-line name/description values and quoting when values
could be interpreted as YAML syntax.
# Testing
Not run (skill-only change.)
Adds a new feature
`enable_request_compression` that will compress using zstd requests to
the codex-backend. Currently only enabled for codex-backend so only enabled for openai providers when using chatgpt::auth even when the feature is enabled
Added a new info log line too for evaluating the compression ratio and
overhead off compressing before requesting. You can enable with
`RUST_LOG=$RUST_LOG,codex_client::transport=info`
```
2026-01-06T00:09:48.272113Z INFO codex_client::transport: Compressed request body with zstd pre_compression_bytes=28914 post_compression_bytes=11485 compression_duration_ms=0
```
We used to override truncation policy by comparing model info vs config
value in context manager. A better way to do it is to construct model
info using the config value
**Summary**
This PR makes “ApprovalDecision::AcceptForSession / don’t ask again this
session” actually work for `apply_patch` approvals by caching approvals
based on absolute file paths in codex-core, properly wiring it through
app-server v2, and exposing the choice in both TUI and TUI2.
- This brings `apply_patch` calls to be at feature-parity with general
shell commands, which also have a "Yes, and don't ask again" option.
- This also fixes VSCE's "Allow this session" button to actually work.
While we're at it, also split the app-server v2 protocol's
`ApprovalDecision` enum so execpolicy amendments are only available for
command execution approvals.
**Key changes**
- Core: per-session patch approval allowlist keyed by absolute file
paths
- Handles multi-file patches and renames/moves by recording both source
and destination paths for `Update { move_path: Some(...) }`.
- Extend the `Approvable` trait and `ApplyPatchRuntime` to work with
multiple keys, because an `apply_patch` tool call can modify multiple
files. For a request to be auto-approved, we will need to check that all
file paths have been approved previously.
- App-server v2: honor AcceptForSession for file changes
- File-change approval responses now map AcceptForSession to
ReviewDecision::ApprovedForSession (no longer downgraded to plain
Approved).
- Replace `ApprovalDecision` with two enums:
`CommandExecutionApprovalDecision` and `FileChangeApprovalDecision`
- TUI / TUI2: expose “don’t ask again for these files this session”
- Patch approval overlays now include a third option (“Yes, and don’t
ask again for these files this session (s)”).
- Snapshot updates for the approval modal.
**Tests added/updated**
- Core:
- Integration test that proves ApprovedForSession on a patch skips the
next patch prompt for the same file
- App-server:
- v2 integration test verifying
FileChangeApprovalDecision::AcceptForSession works properly
**User-visible behavior**
- When the user approves a patch “for session”, future patches touching
only those previously approved file(s) will no longer prompt gain during
that session (both via app-server v2 and TUI/TUI2).
**Manual testing**
Tested both TUI and TUI2 - see screenshots below.
TUI:
<img width="1082" height="355" alt="image"
src="https://github.com/user-attachments/assets/adcf45ad-d428-498d-92fc-1a0a420878d9"
/>
TUI2:
<img width="1089" height="438" alt="image"
src="https://github.com/user-attachments/assets/dd768b1a-2f5f-4bd6-98fd-e52c1d3abd9e"
/>
> // todo(aibrahim): why are we passing model here while it can change?
we update it on each turn with `.with_model`
> //TODO(aibrahim): run CI in release mode.
although it's good to have, release builds take double the time tests
take.
> // todo(aibrahim): make this async function
we figured out another way of doing this sync
- Merge ModelFamily into ModelInfo
- Remove logic for adding instructions to apply patch
- Add compaction limit and visible context window to `ModelInfo`
See https://rustsec.org/advisories/RUSTSEC-2026-0002.
Though our `ratatui` fork has a transitive dep on an older version of
the `lru` crate, so to get CI green ASAP, this PR also adds an exception
to `deny.toml` for `RUSTSEC-2026-0002`, but hopefully this will be
short-lived.
Handle /review <instructions> in the TUI and TUI2 by routing it as a
custom review command instead of plain text, wiring command dispatch and
adding composer coverage so typing /review text starts a review directly
rather than posting a message. User impact: /review with arguments now
kicks off the review flow, previously it would just forward as a plain
command and not actually start a review.
Fixes apply.rs path parsing so
- quoted diff headers are tokenized and extracted correctly,
- /dev/null headers are ignored before prefix stripping to avoid bogus
dev/null paths, and
- git apply output paths are unescaped from C-style quoting.
**Why**
This prevents potentially missed staging and misclassified paths when
applying or reverting patches, which could lead to incorrect behavior
for repos with spaces or escaped characters in filenames.
**Impact**
I checked and this is only used in the cloud tasks support and `codex
apply <task_id>` flow.
Set login=false for the shell tool in the timing-based parallelism test
so it does not depend on slow user login shells, making the test
deterministic without user-facing changes. This prevents occasional
flakes when running locally.
With `config.toml`:
```
model = "gpt-5.1-codex"
```
(where `gpt-5.1-codex` has `show_in_picker: false` in
[`model_presets.rs`](https://github.com/openai/codex/blob/main/codex-rs/core/src/models_manager/model_presets.rs);
this happens if the user hasn't used codex in a while so they didn't see
the popup before their model was changed to `show_in_picker: false`)
The upgrade picker used to not show (because `gpt-5.1-codex` was
filtered out of the model list in code). Now, the filtering is done
downstream in tui and app-server, so the model upgrade popup shows:
<img width="1503" height="227" alt="Screenshot 2026-01-06 at 5 04 37 PM"
src="https://github.com/user-attachments/assets/26144cc2-0b3f-4674-ac17-e476781ec548"
/>
Add `web_search_cached` feature to config. Enables `web_search` tool
with access only to cached/indexed results (see
[docs](https://platform.openai.com/docs/guides/tools-web-search#live-internet-access)).
This takes precedence over the existing `web_search_request`, which
continues to enable `web_search` over live results as it did before.
`web_search_cached` is disabled for review mode, as `web_search_request`
is.
Currently we don't load config properly for app server conversations.
see:
https://linear.app/openai/issue/CODEX-3956/config-flags-not-respected-in-codex-app-server.
This PR fixes that by respecting the config passed in.
Tested by running `cargo build -p codex-cli &&
RUST_LOG=codex_app_server=debug CODEX_BIN=target/debug/codex cargo run
-p codex-app-server-test-client -- \
--config
model_providers.mock_provider.base_url=\"http://localhost:4010/v2\" \
--config model_provider=\"mock_provider\" \
--config model_providers.mock_provider.name="hello" \
send-message-v2 "hello"`
and verified that the mock_provider is called instead of default
provider.
#closes
https://linear.app/openai/issue/CODEX-3956/config-flags-not-respected-in-codex-app-server
---------
Co-authored-by: Michael Bolin <mbolin@openai.com>
Add `thread/rollback` to app-server to support IDEs undo-ing the last N
turns of a thread.
For context, an IDE partner will be supporting an "undo" capability
where the IDE (the app-server client) will be responsible for reverting
the local changes made during the last turn. To support this well, we
also need a way to drop the last turn (or more generally, the last N
turns) from the agent's context. This is what `thread/rollback` does.
**Core idea**: A Thread rollback is represented as a persisted event
message (EventMsg::ThreadRollback) in the rollout JSONL file, not by
rewriting history. On resume, both the model's context (core replay) and
the UI turn list (app-server v2's thread history builder) apply these
markers so the pruned history is consistent across live conversations
and `thread/resume`.
Implementation notes:
- Rollback only affects agent context and appends to the rollout file;
clients are responsible for reverting files on disk.
- If a thread rollback is currently in progress, subsequent
`thread/rollback` calls are rejected.
- Because we use `CodexConversation::submit` and codex core tracks
active turns, returning an error on concurrent rollbacks is communicated
via an `EventMsg::Error` with a new variant
`CodexErrorInfo::ThreadRollbackFailed`. app-server watches for that and
sends the BAD_REQUEST RPC response.
Tests cover thread rollbacks in both core and app-server, including when
`num_turns` > existing turns (which clears all turns).
**Note**: this explicitly does **not** behave like `/undo` which we just
removed from the CLI, which does the opposite of what `thread/rollback`
does. `/undo` reverts local changes via ghost commits/snapshots and does
not modify the agent's context / conversation history.
### Motivation
- Fix a visual bug where transcript text could bleed through the
on-screen copy "pill" overlay.
- Ensure the copy affordance fully covers the underlying buffer so the
pill background is solid and consistent with styling.
- Document the approach in-code to make the background-clearing
rationale explicit.
### Description
- Clear the pill area before drawing by iterating `Rect::positions()`
and calling `cell.set_symbol(" ")` and `cell.set_style(base_style)` in
`render_copy_pill` in `transcript_copy_ui.rs`.
- Added an explanatory comment for why the pill background is explicitly
cleared.
- Added a unit test `copy_pill_clears_background` and committed the
corresponding snapshot file to validate the rendering behavior.
### Testing
- Ran `just fmt` (formatting completed; non-blocking environment warning
may appear).
- Ran `just fix -p codex-tui2` to apply lints/fixes (completed).
- Ran `cargo test -p codex-tui2` and all tests passed (snapshot updated
and tests succeeded).
------
[Codex
Task](https://chatgpt.com/codex/tasks/task_i_695c9b23e9b8832997d5a457c4d83410)
Added an agent control plane that lets sessions spawn or message other
conversations via `AgentControl`.
`AgentBus` (core/src/agent/bus.rs) keeps track of the last known status
of a conversation.
ConversationManager now holds shared state behind an Arc so AgentControl
keeps only a weak back-reference, the goal is just to avoid explicit
cycle reference.
Follow-ups:
* Build a small tool in the TUI to be able to see every agent and send
manual message to each of them
* Handle approval requests in this TUI
* Add tools to spawn/communicate between agents (see related design)
* Define agent types
Force an announcement tooltip in the CLI. This query the gh repo on this
[file](https://raw.githubusercontent.com/openai/codex/main/announcement_tip.toml)
which contains announcements in TOML looking like this:
```
# Example announcement tips for Codex TUI.
# Each [[announcements]] entry is evaluated in order; the last matching one is shown.
# Dates are UTC, formatted as YYYY-MM-DD. The from_date is inclusive and the to_date is exclusive.
# version_regex matches against the CLI version (env!("CARGO_PKG_VERSION")); omit to apply to all versions.
# target_app specify which app should display the announcement (cli, vsce, ...).
[[announcements]]
content = "Welcome to Codex! Check out the new onboarding flow."
from_date = "2024-10-01"
to_date = "2024-10-15"
version_regex = "^0\\.0\\.0$"
target_app = "cli"
```
To make this efficient, the announcement is queried on a best effort
basis at the launch of the CLI (no refresh made after this).
This is done in an async way and we display the announcement (with 100%
probability) iff the announcement is available, the cache is correctly
warmed and there is a matching announcement (matching is recomputed for
each new session).
Fixes ReadinessFlag::subscribe to avoid handing out token 0 or duplicate
tokens on i32 wrap-around, adds regression tests, and prevents readiness
gates from getting stuck waiting on an unmarkable or mis-authorized
token.
Background
Streaming assistant prose in tui2 was being rendered with viewport-width
wrapping during streaming, then stored in history cells as already split
`Line`s. Those width-derived breaks became indistinguishable from hard
newlines, so the transcript could not "un-split" on resize. This also
degraded copy/paste, since soft wraps looked like hard breaks.
What changed
- Introduce width-agnostic `MarkdownLogicalLine` output in
`tui2/src/markdown_render.rs`, preserving markdown wrap semantics:
initial/subsequent indents, per-line style, and a preformatted flag.
- Update the streaming collector (`tui2/src/markdown_stream.rs`) to emit
logical lines (newline-gated) and remove any captured viewport width.
- Update streaming orchestration (`tui2/src/streaming/*`) to queue and
emit logical lines, producing `AgentMessageCell::new_logical(...)`.
- Make `AgentMessageCell` store logical lines and wrap at render time in
`HistoryCell::transcript_lines_with_joiners(width)`, emitting joiners so
copy/paste can join soft-wrap continuations correctly.
Overlay deferral
When an overlay is active, defer *cells* (not rendered `Vec<Line>`) and
render them at overlay close time. This avoids baking width-derived
wraps based on a stale width.
Tests + docs
- Add resize/reflow regression tests + snapshots for streamed agent
output.
- Expand module/API docs for the new logical-line streaming pipeline and
clarify joiner semantics.
- Align scrollback-related docs/comments with current tui2 behavior
(main draw loop does not flush queued "history lines" to the terminal).
More details
See `codex-rs/tui2/docs/streaming_wrapping_design.md` for the full
problem statement and solution approach, and
`codex-rs/tui2/docs/tui_viewport_and_history.md` for viewport vs printed
output behavior.
Trim whitespace when validating '*** Begin Patch'/'*** End Patch'
markers in codex-apply-patch so padded marker lines parse as intended,
and add regression coverage (unit + fixture scenario); this avoids
apply_patch failures when models include extra spacing. Tested with
cargo test -p codex-apply-patch.
**Motivation**
- Bring `codex exec resume` to parity with top‑level flags so global
options (git check bypass, json, model, sandbox toggles) work after the
subcommand, including when outside a git repo.
**Description**
- Exec CLI: mark `--skip-git-repo-check`, `--json`, `--model`,
`--full-auto`, and `--dangerously-bypass-approvals-and-sandbox` as
global so they’re accepted after `resume`.
- Tests: add `exec_resume_accepts_global_flags_after_subcommand` to
verify those flags work when passed after `resume`.
**Testing**
- `just fmt`
- `cargo test -p codex-exec` (pass; ran with elevated perms to allow
network/port binds)
- Manual: exercised `codex exec resume` with global flags after the
subcommand to confirm behavior.
We've seen reports that people who try to login on a remote/headless
machine will open the login link on their own machine and got errors.
Update the instructions to ask those users to use `codex login
--device-auth` instead.
<img width="1434" height="938" alt="CleanShot 2026-01-05 at 11 35 02@2x"
src="https://github.com/user-attachments/assets/2b209953-6a42-4eb0-8b55-bb0733f2e373"
/>
Adds an optional `justification` parameter to the `prefix_rule()`
execpolicy DSL so policy authors can attach human-readable rationale to
a rule. That justification is propagated through parsing/matching and
can be surfaced to the model (or approval UI) when a command is blocked
or requires approval.
When a command is rejected (or gated behind approval) due to policy, a
generic message makes it hard for the model/user to understand what went
wrong and what to do instead. Allowing policy authors to supply a short
justification improves debuggability and helps guide the model toward
compliant alternatives.
Example:
```python
prefix_rule(
pattern = ["git", "push"],
decision = "forbidden",
justification = "pushing is blocked in this repo",
)
```
If Codex tried to run `git push origin main`, now the failure would
include:
```
`git push origin main` rejected: pushing is blocked in this repo
```
whereas previously, all it was told was:
```
execpolicy forbids this command
```
The elevated sandbox creates two new Windows users - CodexSandboxOffline
and CodexSandboxOnline. This is necessary, so this PR does all that it
can to "hide" those users. It uses the registry plus directory flags (on
their home directories) to get them to show up as little as possible.
