The underlying issue is that when we encountered an error starting a
conversation (any sort of error, though making `$CODEX_HOME/rules` a
file rather than folder was the example in #8803), then we were writing
the message to stderr, but this could be printed over by our UI
framework so the user would not see it. In general, we disallow the use
of `eprintln!()` in this part of the code for exactly this reason,
though this was suppressed by an `#[allow(clippy::print_stderr)]`.
This attempts to clean things up by changing `handle_event()` and
`handle_tui_event()` to return a `Result<AppRunControl>` instead of a
`Result<bool>`, which is a new type introduced in this PR (and depends
on `ExitReason`, also a new type):
```rust
#[derive(Debug)]
pub(crate) enum AppRunControl {
Continue,
Exit(ExitReason),
}
#[derive(Debug, Clone)]
pub enum ExitReason {
UserRequested,
Fatal(String),
}
```
This makes it possible to exit the primary control flow of the TUI with
richer information. This PR adds `ExitReason` to the existing
`AppExitInfo` struct and updates `handle_app_exit()` to print the error
and exit code `1` in the event of `ExitReason::Fatal`.
I tried to create an integration test for this, but it was a bit
involved, so I published it as a separate PR:
https://github.com/openai/codex/pull/9166. For this PR, please have
faith in my manual testing!
Fixes https://github.com/openai/codex/issues/8803.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/9011).
* #9166
* __->__ #9011
## **Problem**
Codex’s TUI uses a single “task running” indicator (spinner + Esc interrupt hint)
to communicate “the UI is busy”. In practice, “busy” can mean two different
things: an agent turn is running, or MCP servers are still starting up. Without a
clear contract, those lifecycles can interfere: startup completion can clear the
spinner while a turn is still in progress, or the UI can appear idle while MCP is
still booting. This is user-visible confusion during the most important moments
(startup and the first turn), so it was worth making the contract explicit and
guarding it.
## **Mental model**
`ChatWidget` is the UI-side adapter for the `codex_core::protocol` event stream.
It receives `EventMsg` events and updates two major UI surfaces: the transcript
(history/streaming cells) and the bottom pane (composer + status indicator).
The key concept after this change is that the bottom pane’s “task running”
indicator is treated as **derived UI-busy state**, not “agent is running”. It is
considered active while either:
- an agent turn is in progress (`TurnStarted` → completion/abort), or
- MCP startup is in progress (`McpStartupUpdate` → `McpStartupComplete`).
Those lifecycles are tracked independently, and the bottom-pane indicator is
defined as their union.
## **Non-goals**
- This does not introduce separate UI indicators for “turn busy” vs “MCP busy”.
- This does not change MCP startup behavior, ordering guarantees, or core
protocol semantics.
- This does not rework unrelated status/header rendering or transcript layout.
## **Tradeoffs**
- The “one flag represents multiple lifecycles” approach remains lossy: it
preserves correct “busy vs idle” semantics but cannot express *which* kind of
busy is happening without further UI changes.
- The design keeps complexity low by keeping a single derived boolean, rather
than adding a more expressive bottom-pane state machine. That’s chosen because
it matches existing UX and minimizes churn while fixing the confusion.
## **Architecture**
- `codex-core` owns the actual lifecycles and emits `codex_core::protocol`
events.
- `ChatWidget` owns the UI interpretation of those lifecycles. It is responsible
for keeping the bottom pane’s derived “busy” state consistent with the event
stream, and for updating the status header when MCP progress updates arrive.
- The bottom pane remains a dumb renderer of the single “task running” flag; it
does not learn about MCP or agent turns directly.
## **Observability**
- When working: the spinner/Esc hint stays visible during MCP startup and does
not disappear mid-turn when `McpStartupComplete` arrives; startup status
headers can update without clearing “busy” for an active turn.
- When broken: you’ll see the spinner/hint flicker off while output is still
streaming, or the UI appears idle while MCP startup status is still changing.
## **Tests**
- Adds/strengthens a regression test that asserts MCP startup completion does
not clear the “task running” indicator for an active turn (in both `tui` and
`tui2` variants).
- These tests prove the **contract** (“busy is the union of turn + startup”) at
the UI boundary; they do not attempt to validate MCP startup ordering,
real-world startup timing, or backend integration behavior.
Fixes#7017
Signed-off-by: 2mawi2 <2mawi2@users.noreply.github.com>
Co-authored-by: 2mawi2 <2mawi2@users.noreply.github.com>
Co-authored-by: Josh McKinney <joshka@openai.com>
Sending a message during /review interrupts the review, whereas during
normal operation, sending a message while the agent is running will
queue the message. This is unexpected behavior, and since /review
usually takes a while, it takes away a potentially useful operation.
Summary
- Treat review mode as an active task for message queuing so inputs
don’t inject into the running review turn.
- Prevents user submissions from rendering immediately in the transcript
while the review continues streaming.
- Keeps review UX consistent with normal “task running” behavior and
avoids accidental interrupt/replacement.
Notes
- This change only affects UI queuing logic; core review flow and task
lifecycle remain unchanged.
### Problem
Ctrl+T transcript overlay can omit in-flight coalesced tool calls because it
renders only committed transcript cells while the main viewport can render the
current in-flight ChatWidget.active_cell immediately.
### Mental model
The UI has both committed transcript cells (finalized HistoryCell entries) and
an in-flight active cell that can mutate in place while streaming, often
representing a coalesced exec/tool group. The transcript overlay renders
committed cells plus a render-only live tail derived from the current active
cell. The live tail is cached and only recomputed when its cache key changes,
which is derived from terminal width (wrapping), active-cell revision
(in-place mutations), stream continuation (spacing), and animation tick
(time-based visuals).
### Non-goals
This does not change coalescing rules, flush boundaries, or when active cells
become committed. It does not change tool-call semantics or transcript
persistence; it is a rendering-only improvement for the overlay.
### Tradeoffs
This adds cache invalidation complexity: correctness depends on bumping an
active-cell revision (and/or providing an animation tick) when the active cell
mutates in place. The mechanism is implemented in both codex-tui and codex-tui2,
which keeps behavior consistent but risks drift if future changes are not
applied in lockstep.
### Architecture
App special-cases transcript overlay draws to sync a live tail from ChatWidget
into TranscriptOverlay. TranscriptOverlay remains the owner of committed
transcript cells; the live tail is an optional appended renderable.
HistoryCell::transcript_animation_tick() allows time-dependent transcript output
(spinner/shimmer) to invalidate the cached tail without requiring data mutation.
### Observability
Manual verification is to open Ctrl+T while an exploring/coalesced active cell
is still in-flight and confirm the overlay includes the same in-flight tool-call
group the main viewport shows. The overlay is kept in sync by App passing an
active-cell key and transcript lines into TranscriptOverlay::sync_live_tail; the
key must change when the active cell mutates or animates.
### Tests
Snapshot tests validate that the transcript overlay renders a live tail appended
after committed cells and that identical keys short-circuit recomputation. Unit
tests validate that active-cell revision bumps occur on specific in-place
mutations (e.g. unified exec wait cell command display becoming known late) so
cached tails are invalidated.
## Documentation patches (module, type, function)
### Module-level docs (invariants + mechanisms)
- codex-rs/tui/src/app_backtrack.rs:1
- codex-rs/tui/src/chatwidget.rs:1
- codex-rs/tui/src/pager_overlay.rs:1
- codex-rs/tui/src/history_cell.rs:1
- codex-rs/tui2/src/app_backtrack.rs:1
- codex-rs/tui2/src/chatwidget.rs:1
- codex-rs/tui2/src/pager_overlay.rs:1
- codex-rs/tui2/src/history_cell.rs:1
### Type-level docs (cache key + invariants)
- codex-rs/tui/src/chatwidget.rs (ChatWidget.active_cell_revision, ActiveCellTranscriptKey)
- codex-rs/tui/src/pager_overlay.rs (TranscriptOverlay live tail storage model)
- codex-rs/tui/src/history_cell.rs (HistoryCell::transcript_animation_tick, UnifiedExecWaitCell::update_command_display)
- Mirrored in codex-rs/tui2/src/chatwidget.rs, codex-rs/tui2/src/pager_overlay.rs, codex-rs/tui2/src/history_cell.rs
### Function-level docs (why/when/guarantees/pitfalls)
- codex-rs/tui/src/app_backtrack.rs (overlay_forward_event)
- codex-rs/tui/src/chatwidget.rs (active_cell_transcript_key, active_cell_transcript_lines)
- codex-rs/tui/src/pager_overlay.rs (sync_live_tail, take_live_tail_renderable)
- codex-rs/tui/src/history_cell.rs (transcript_animation_tick, UnifiedExecWaitCell::update_command_display)
- Mirrored in codex-rs/tui2 equivalents where present
### Validation performed
- cd codex-rs && just fmt
- cd codex-rs && cargo test -p codex-tui
- cd codex-rs && cargo test -p codex-tui2
## Design inconsistencies / risks
- Cache invalidation is a distributed responsibility: any future in-place active
cell transcript mutation that forgets to bump active_cell_revision (or expose
an animation tick) can leave the transcript overlay live tail out of sync with
the main viewport.
- TranscriptOverlay tail handling assumes a structural invariant that the live
tail, when present, is exactly one trailing renderable after the committed cell
renderables; if renderable construction changes in a way that violates that
assumption, tail insertion/removal logic becomes incorrect.
- codex-tui and codex-tui2 duplicate the live-tail mechanism; the documentation
is aligned, but the implementation can still drift unless changes continue to
be applied in lockstep.
Agent wouldn't "see" attached images and would instead try to use the
view_file tool:
<img width="1516" height="504" alt="image"
src="https://github.com/user-attachments/assets/68a705bb-f962-4fc1-9087-e932a6859b12"
/>
In this PR, we wrap image content items in XML tags with the name of
each image (now just a numbered name like `[Image #1]`), so that the
model can understand inline image references (based on name). We also
put the image content items above the user message which the model seems
to prefer (maybe it's more used to definitions being before references).
We also tweak the view_file tool description which seemed to help a bit
Results on a simple eval set of images:
Before
<img width="980" height="310" alt="image"
src="https://github.com/user-attachments/assets/ba838651-2565-4684-a12e-81a36641bf86"
/>
After
<img width="918" height="322" alt="image"
src="https://github.com/user-attachments/assets/10a81951-7ee6-415e-a27e-e7a3fd0aee6f"
/>
```json
[
{
"id": "single_describe",
"prompt": "Describe the attached image in one sentence.",
"images": ["image_a.png"]
},
{
"id": "single_color",
"prompt": "What is the dominant color in the image? Answer with a single color word.",
"images": ["image_b.png"]
},
{
"id": "orientation_check",
"prompt": "Is the image portrait or landscape? Answer in one sentence.",
"images": ["image_c.png"]
},
{
"id": "detail_request",
"prompt": "Look closely at the image and call out any small details you notice.",
"images": ["image_d.png"]
},
{
"id": "two_images_compare",
"prompt": "I attached two images. Are they the same or different? Briefly explain.",
"images": ["image_a.png", "image_b.png"]
},
{
"id": "two_images_captions",
"prompt": "Provide a short caption for each image (Image 1, Image 2).",
"images": ["image_c.png", "image_d.png"]
},
{
"id": "multi_image_rank",
"prompt": "Rank the attached images from most colorful to least colorful.",
"images": ["image_a.png", "image_b.png", "image_c.png"]
},
{
"id": "multi_image_choice",
"prompt": "Which image looks more vibrant? Answer with 'Image 1' or 'Image 2'.",
"images": ["image_b.png", "image_d.png"]
}
]
```
Elevated Sandbox NUX:
* prompt for elevated sandbox setup when agent mode is selected (via
/approvals or at startup)
* prompt for degraded sandbox if elevated setup is declined or fails
* introduce /elevate-sandbox command to upgrade from degraded
experience.
**Before:**
```
Error loading configuration: value `Never` is not in the allowed set [OnRequest]
```
**After:**
```
Error loading configuration: invalid value for `approval_policy`: `Never` is not in the
allowed set [OnRequest] (set by MDM com.openai.codex:requirements_toml_base64)
```
Done by introducing a new struct `ConfigRequirementsWithSources` onto
which we `merge_unset_fields` now. Also introduces a pair of requirement
value and its `RequirementSource` (inspired by `ConfigLayerSource`):
```rust
pub struct Sourced<T> {
pub value: T,
pub source: RequirementSource,
}
```
**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"
/>
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"
/>
### What
Builds on #8293.
Add `additional_details`, which contains the upstream error message, to
relevant structures used to pass along retryable `StreamError`s.
Uses the new TUI status indicator's `details` field (shows under the
status header) to display the `additional_details` error to the user on
retryable `Reconnecting...` errors. This adds clarity for users for
retryable errors.
Will make corresponding change to VSCode extension to show
`additional_details` as expandable from the `Reconnecting...` cell.
Examples:
<img width="1012" height="326" alt="image"
src="https://github.com/user-attachments/assets/f35e7e6a-8f5e-4a2f-a764-358101776996"
/>
<img width="1526" height="358" alt="image"
src="https://github.com/user-attachments/assets/0029cbc0-f062-4233-8650-cc216c7808f0"
/>
Add `ctrl+g` shortcut to enable opening current prompt in configured
editor (`$VISUAL` or `$EDITOR`).
- Prompt is updated with editor's content upon editor close.
- Paste placeholders are automatically expanded when opening the
external editor, and are not "recompressed" on close
- They could be preserved in the editor, but it would be hard to prevent
the user from modifying the placeholder text directly, which would drop
the mapping to the `pending_paste` value
- Image placeholders stay as-is
- `ctrl+g` explanation added to shortcuts menu, snapshot tests updated
https://github.com/user-attachments/assets/4ee05c81-fa49-4e99-8b07-fc9eef0bbfce
# External (non-OpenAI) Pull Request Requirements
Before opening this Pull Request, please read the dedicated
"Contributing" markdown file or your PR may be closed:
https://github.com/openai/codex/blob/main/docs/contributing.md
If your PR conforms to our contribution guidelines, replace this text
with a detailed and high quality description of your changes.
Include a link to a bug report or enhancement request.
This will make it easier to test for expected errors in unit tests since
we can compare based on the field values rather than the message (which
might change over time). See https://github.com/openai/codex/pull/8298
for an example.
It also ensures more consistency in the way a `ConstraintError` is
constructed.
https://github.com/openai/codex/pull/8235 introduced `ConfigBuilder` and
this PR updates all call non-test call sites to use it instead of
`Config::load_from_base_config_with_overrides()`.
This is important because `load_from_base_config_with_overrides()` uses
an empty `ConfigRequirements`, which is a reasonable default for testing
so the tests are not influenced by the settings on the host. This method
is now guarded by `#[cfg(test)]` so it cannot be used by business logic.
Because `ConfigBuilder::build()` is `async`, many of the test methods
had to be migrated to be `async`, as well. On the bright side, this made
it possible to eliminate a bunch of `block_on_future()` stuff.
# External (non-OpenAI) Pull Request Requirements
Before opening this Pull Request, please read the dedicated
"Contributing" markdown file or your PR may be closed:
https://github.com/openai/codex/blob/main/docs/contributing.md
If your PR conforms to our contribution guidelines, replace this text
with a detailed and high quality description of your changes.
Include a link to a bug report or enhancement request.
# External (non-OpenAI) Pull Request Requirements
Before opening this Pull Request, please read the dedicated
"Contributing" markdown file or your PR may be closed:
https://github.com/openai/codex/blob/main/docs/contributing.md
If your PR conforms to our contribution guidelines, replace this text
with a detailed and high quality description of your changes.
Include a link to a bug report or enhancement request.
Constrain `approval_policy` through new `admin_policy` config.
This PR will:
1. Add a `admin_policy` section to config, with a single field (for now)
`allowed_approval_policies`. This list constrains the set of
user-settable `approval_policy`s.
2. Introduce a new `Constrained<T>` type, which combines a current value
and a validator function. The validator function ensures disallowed
values are not set.
3. Change the type of `approval_policy` on `Config` and
`SessionConfiguration` from `AskForApproval` to
`Constrained<AskForApproval>`. The validator function is set by the
values passed into `allowed_approval_policies`.
4. `GenericDisplayRow`: add a `disabled_reason: Option<String>`. When
set, it disables selection of the value and indicates as such in the
menu. This also makes it unselectable with arrow keys or numbers. This
is used in the `/approvals` menu.
Follow ups are:
1. Do the same thing to `sandbox_policy`.
2. Propagate the allowed set of values through app-server for the
extension (though already this should prevent app-server from setting
this values, it's just that we want to disable UI elements that are
unsettable).
Happy to split this PR up if you prefer, into the logical numbered areas
above. Especially if there are parts we want to gavel on separately
(e.g. admin_policy).
Disabled full access:
<img width="1680" height="380" alt="image"
src="https://github.com/user-attachments/assets/1fb61c8c-1fcb-4dc4-8355-2293edb52ba0"
/>
Disabled `--yolo` on startup:
<img width="749" height="76" alt="image"
src="https://github.com/user-attachments/assets/0a1211a0-6eb1-40d6-a1d7-439c41e94ddb"
/>
CODEX-4087
refactor the way we load and manage skills:
1. Move skill discovery/caching into SkillsManager and reuse it across
sessions.
2. Add the skills/list API (Op::ListSkills/SkillsListResponse) to fetch
skills for one or more cwds. Also update app-server for VSCE/App;
3. Trigger skills/list during session startup so UIs preload skills and
handle errors immediately.
Changes the `writable_roots` field of the `WorkspaceWrite` variant of
the `SandboxPolicy` enum from `Vec<PathBuf>` to `Vec<AbsolutePathBuf>`.
This is helpful because now callers can be sure the value is an absolute
path rather than a relative one. (Though when using an absolute path in
a Seatbelt config policy, we still have to _canonicalize_ it first.)
Because `writable_roots` can be read from a config file, it is important
that we are able to resolve relative paths properly using the parent
folder of the config file as the base path.
## Problem
The introduction of `notify_sandbox_state_change()` in #7112 caused a
regression where the blocking call in `Session::new()` waits for all MCP
servers to fully initialize before returning. This prevents the TUI
event loop from starting, resulting in `McpStartupUpdateEvent` messages
being emitted but never consumed or displayed. As a result, the app
appears to hang during startup, and users do not see the expected
"Booting MCP server: {name}" status line.
Issue: [#7827](https://github.com/openai/codex/issues/7827)
## Solution
This change moves sandbox state notification into each MCP server's
background initialization task. The notification is sent immediately
after the server transitions to the Ready state. This approach:
- Avoids blocking `Session::new()`, allowing the TUI event loop to start
promptly.
- Ensures each MCP server receives its sandbox state before handling any
tool calls.
- Restores the display of "Booting MCP server" status lines during
startup.
## Key Changes
- Added `ManagedClient::notify_sandbox_state()` method.
- Passed sandbox_state to `McpConnectionManager::initialize()`.
- Sends sandbox state notification in the background task after the
server reaches Ready status.
- Removed blocking notify_sandbox_state_change() methods.
- Added a chatwidget snapshot test for the "Booting MCP server" status
line.
## Regression Details
Regression was bisected to #7112, which introduced the blocking
behavior.
---------
Co-authored-by: Michael Bolin <bolinfest@gmail.com>
Co-authored-by: Michael Bolin <mbolin@openai.com>
1. Skills load once in core at session start; the cached outcome is
reused across core and surfaced to TUI via SessionConfigured.
2. TUI detects explicit skill selections, and core injects the matching
SKILL.md content into the turn when a selected skill is present.
- Make Config.model optional and centralize default-selection logic in
ModelsManager, including a default_model helper (with
codex-auto-balanced when available) so sessions now carry an explicit
chosen model separate from the base config.
- Resolve `model` once in `core` and `tui` from config. Then store the
state of it on other structs.
- Move refreshing models to be before resolving the default model
