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"
/>
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.
What changed
- Added `outputSchema` support to the app-server APIs, mirroring `codex
exec --output-schema` behavior.
- V1 `sendUserTurn` now accepts `outputSchema` and constrains the final
assistant message for that turn.
- V2 `turn/start` now accepts `outputSchema` and constrains the final
assistant message for that turn (explicitly per-turn only).
Core behavior
- `Op::UserTurn` already supported `final_output_json_schema`; now V1
`sendUserTurn` forwards `outputSchema` into that field.
- `Op::UserInput` now carries `final_output_json_schema` for per-turn
settings updates; core maps it into
`SessionSettingsUpdate.final_output_json_schema` so it applies to the
created turn context.
- V2 `turn/start` does NOT persist the schema via `OverrideTurnContext`
(it’s applied only for the current turn). Other overrides
(cwd/model/etc) keep their existing persistent behavior.
API / docs
- `codex-rs/app-server-protocol/src/protocol/v1.rs`: add `output_schema:
Option<serde_json::Value>` to `SendUserTurnParams` (serialized as
`outputSchema`).
- `codex-rs/app-server-protocol/src/protocol/v2.rs`: add `output_schema:
Option<JsonValue>` to `TurnStartParams` (serialized as `outputSchema`).
- `codex-rs/app-server/README.md`: document `outputSchema` for
`turn/start` and clarify it applies only to the current turn.
- `codex-rs/docs/codex_mcp_interface.md`: document `outputSchema` for v1
`sendUserTurn` and v2 `turn/start`.
Tests added/updated
- New app-server integration tests asserting `outputSchema` is forwarded
into outbound `/responses` requests as `text.format`:
- `codex-rs/app-server/tests/suite/output_schema.rs`
- `codex-rs/app-server/tests/suite/v2/output_schema.rs`
- Added per-turn semantics tests (schema does not leak to the next
turn):
- `send_user_turn_output_schema_is_per_turn_v1`
- `turn_start_output_schema_is_per_turn_v2`
- Added protocol wire-compat tests for the merged op:
- serialize omits `final_output_json_schema` when `None`
- deserialize works when field is missing
- serialize includes `final_output_json_schema` when `Some(schema)`
Call site updates (high level)
- Updated all `Op::UserInput { .. }` constructions to include
`final_output_json_schema`:
- `codex-rs/app-server/src/codex_message_processor.rs`
- `codex-rs/core/src/codex_delegate.rs`
- `codex-rs/mcp-server/src/codex_tool_runner.rs`
- `codex-rs/tui/src/chatwidget.rs`
- `codex-rs/tui2/src/chatwidget.rs`
- plus impacted core tests.
Validation
- `just fmt`
- `cargo test -p codex-core`
- `cargo test -p codex-app-server`
- `cargo test -p codex-mcp-server`
- `cargo test -p codex-tui`
- `cargo test -p codex-tui2`
- `cargo test -p codex-protocol`
- `cargo clippy --all-features --tests --profile dev --fix -- -D
warnings`
### 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"
/>
This PR introduces a `codex-utils-cargo-bin` utility crate that
wraps/replaces our use of `assert_cmd::Command` and
`escargot::CargoBuild`.
As you can infer from the introduction of `buck_project_root()` in this
PR, I am attempting to make it possible to build Codex under
[Buck2](https://buck2.build) as well as `cargo`. With Buck2, I hope to
achieve faster incremental local builds (largely due to Buck2's
[dice](https://buck2.build/docs/insights_and_knowledge/modern_dice/)
build strategy, as well as benefits from its local build daemon) as well
as faster CI builds if we invest in remote execution and caching.
See
https://buck2.build/docs/getting_started/what_is_buck2/#why-use-buck2-key-advantages
for more details about the performance advantages of Buck2.
Buck2 enforces stronger requirements in terms of build and test
isolation. It discourages assumptions about absolute paths (which is key
to enabling remote execution). Because the `CARGO_BIN_EXE_*` environment
variables that Cargo provides are absolute paths (which
`assert_cmd::Command` reads), this is a problem for Buck2, which is why
we need this `codex-utils-cargo-bin` utility.
My WIP-Buck2 setup sets the `CARGO_BIN_EXE_*` environment variables
passed to a `rust_test()` build rule as relative paths.
`codex-utils-cargo-bin` will resolve these values to absolute paths,
when necessary.
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/8496).
* #8498
* __->__ #8496
This isn't very useful parameter.
logic:
```
if model puts `**` in their reasoning, trim it and visualize the header.
if couldn't trim: don't render
if model doesn't support: don't render
```
We can simplify to:
```
if could trim, visualize header.
if not, don't render
```
This adds logic to load `/etc/codex/config.toml` and associate it with
`ConfigLayerSource::System` on UNIX. I refactored the code so it shares
logic with the creation of the `ConfigLayerSource::User` layer.
### Summary
With codesigning on Mac, Windows and Linux, we should be able to safely
remove `features.rmcp_client` and `use_experimental_use_rmcp_client`
check from the codebase now.
# 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 adds support for `allowed_sandbox_modes` in `requirements.toml` and
provides legacy support for constraining sandbox modes in
`managed_config.toml`. This is converted to `Constrained<SandboxPolicy>`
in `ConfigRequirements` and applied to `Config` such that constraints
are enforced throughout the harness.
Note that, because `managed_config.toml` is deprecated, we do not add
support for the new `external-sandbox` variant recently introduced in
https://github.com/openai/codex/pull/8290. As noted, that variant is not
supported in `config.toml` today, but can be configured programmatically
via app server.
## Summary
- centralize file name derivation in codex-file-search
- reuse the helper in app-server fuzzy search to avoid duplicate logic
- add unit tests for file_name_from_path
## Testing
- cargo test -p codex-file-search
- cargo test -p codex-app-server
## Description
Introduced `ExternalSandbox` policy to cover use case when sandbox
defined by outside environment, effectively it translates to
`SandboxMode#DangerFullAccess` for file system (since sandbox configured
on container level) and configurable `network_access` (either Restricted
or Enabled by outside environment).
as example you can configure `ExternalSandbox` policy as part of
`sendUserTurn` v1 app_server API:
```
{
"conversationId": <id>,
"cwd": <cwd>,
"approvalPolicy": "never",
"sandboxPolicy": {
"type": ""external-sandbox",
"network_access": "enabled"/"restricted"
},
"model": <model>,
"effort": <effort>,
....
}
```
This is a significant change to how layers of configuration are applied.
In particular, the `ConfigLayerStack` now has two important fields:
- `layers: Vec<ConfigLayerEntry>`
- `requirements: ConfigRequirements`
We merge `TomlValue`s across the layers, but they are subject to
`ConfigRequirements` before creating a `Config`.
How I would review this PR:
- start with `codex-rs/app-server-protocol/src/protocol/v2.rs` and note
the new variants added to the `ConfigLayerSource` enum:
`LegacyManagedConfigTomlFromFile` and `LegacyManagedConfigTomlFromMdm`
- note that `ConfigLayerSource` now has a `precedence()` method and
implements `PartialOrd`
- `codex-rs/core/src/config_loader/layer_io.rs` is responsible for
loading "admin" preferences from `/etc/codex/managed_config.toml` and
MDM. Because `/etc/codex/managed_config.toml` is now deprecated in favor
of `/etc/codex/requirements.toml` and `/etc/codex/config.toml`, we now
include some extra information on the `LoadedConfigLayers` returned in
`layer_io.rs`.
- `codex-rs/core/src/config_loader/mod.rs` has major changes to
`load_config_layers_state()`, which is what produces `ConfigLayerStack`.
The docstring has the new specification and describes the various layers
that will be loaded and the precedence order.
- It uses the information from `LoaderOverrides` "twice," both in the
spirit of legacy support:
- We use one instances to derive an instance of `ConfigRequirements`.
Currently, the only field in `managed_config.toml` that contributes to
`ConfigRequirements` is `approval_policy`. This PR introduces
`Constrained::allow_only()` to support this.
- We use a clone of `LoaderOverrides` to derive
`ConfigLayerSource::LegacyManagedConfigTomlFromFile` and
`ConfigLayerSource::LegacyManagedConfigTomlFromMdm` layers, as
appropriate. As before, this ends up being a "best effort" at enterprise
controls, but is enforcement is not guaranteed like it is for
`ConfigRequirements`.
- Now we only create a "user" layer if `$CODEX_HOME/config.toml` exists.
(Previously, a user layer was always created for `ConfigLayerStack`.)
- Similarly, we only add a "session flags" layer if there are CLI
overrides.
- `config_loader/state.rs` contains the updated implementation for
`ConfigLayerStack`. Note the public API is largely the same as before,
but the implementation is quite different. We leverage the fact that
`ConfigLayerSource` is now `PartialOrd` to ensure layers are in the
correct order.
- A `Config` constructed via `ConfigBuilder.build()` will use
`load_config_layers_state()` to create the `ConfigLayerStack` and use
the associated `ConfigRequirements` when constructing the `Config`
object.
- That said, a `Config` constructed via
`Config::load_from_base_config_with_overrides()` does _not_ yet use
`ConfigBuilder`, so it creates a `ConfigRequirements::default()` instead
of loading a proper `ConfigRequirements`. I will fix this in a
subsequent PR.
Then the following files are mostly test changes:
```
codex-rs/app-server/tests/suite/v2/config_rpc.rs
codex-rs/core/src/config/service.rs
codex-rs/core/src/config_loader/tests.rs
```
Again, because we do not always include "user" and "session flags"
layers when the contents are empty, `ConfigLayerStack` sometimes has
fewer layers than before (and the precedence order changed slightly),
which is the main reason integration tests changed.
# 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.
1. Remove PUBLIC skills and introduce SYSTEM skills embedded in the
binary and installed into $CODEX_HOME/skills/.system at startup.
2. Skills are now always enabled (feature flag removed).
3. Update skills/list to accept forceReload and plumb it through (not
used by clients yet).
This PR does various types of cleanup before I can proceed with more
ambitious changes to config loading.
First, I noticed duplicated code across these two methods:
774bd9e432/codex-rs/core/src/config/mod.rs (L314-L324)774bd9e432/codex-rs/core/src/config/mod.rs (L334-L344)
This has now been consolidated in
`load_config_as_toml_with_cli_overrides()`.
Further, I noticed that `Config::load_with_cli_overrides()` took two
similar arguments:
774bd9e432/codex-rs/core/src/config/mod.rs (L308-L311)
The difference between `cli_overrides` and `overrides` was not
immediately obvious to me. At first glance, it appears that one should
be able to be expressed in terms of the other, but it turns out that
some fields of `ConfigOverrides` (such as `cwd` and
`codex_linux_sandbox_exe`) are, by design, not configurable via a
`.toml` file or a command-line `--config` flag.
That said, I discovered that many callers of
`Config::load_with_cli_overrides()` were passing
`ConfigOverrides::default()` for `overrides`, so I created two separate
methods:
- `Config::load_with_cli_overrides(cli_overrides: Vec<(String,
TomlValue)>)`
- `Config::load_with_cli_overrides_and_harness_overrides(cli_overrides:
Vec<(String, TomlValue)>, harness_overrides: ConfigOverrides)`
The latter has a long name, as it is _not_ what should be used in the
common case, so the extra typing is designed to draw attention to this
fact. I tried to update the existing callsites to use the shorter name,
where possible.
Further, in the cases where `ConfigOverrides` is used, usually only a
limited subset of fields are actually set, so I updated the declarations
to leverage `..Default::default()` where possible.
# 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.
### Summary
* Make `app_server.list_models` to be non-blocking and consumers (i.e.
extension) can manage the flow themselves.
* Force config to use remote models and therefore fetch codex-auto model
list.
This attempts to tighten up the types related to "config layers."
Currently, `ConfigLayerEntry` is defined as follows:
bef36f4ae7/codex-rs/core/src/config_loader/state.rs (L19-L25)
but the `source` field is a bit of a lie, as:
- for `ConfigLayerName::Mdm`, it is
`"com.openai.codex/config_toml_base64"`
- for `ConfigLayerName::SessionFlags`, it is `"--config"`
- for `ConfigLayerName::User`, it is `"config.toml"` (just the file
name, not the path to the `config.toml` on disk that was read)
- for `ConfigLayerName::System`, it seems like it is usually
`/etc/codex/managed_config.toml` in practice, though on Windows, it is
`%CODEX_HOME%/managed_config.toml`:
bef36f4ae7/codex-rs/core/src/config_loader/layer_io.rs (L84-L101)
All that is to say, in three out of the four `ConfigLayerName`, `source`
is a `PathBuf` that is not an absolute path (or even a true path).
This PR tries to uplevel things by eliminating `source` from
`ConfigLayerEntry` and turning `ConfigLayerName` into a disjoint union
named `ConfigLayerSource` that has the appropriate metadata for each
variant, favoring the use of `AbsolutePathBuf` where appropriate:
```rust
pub enum ConfigLayerSource {
/// Managed preferences layer delivered by MDM (macOS only).
#[serde(rename_all = "camelCase")]
#[ts(rename_all = "camelCase")]
Mdm { domain: String, key: String },
/// Managed config layer from a file (usually `managed_config.toml`).
#[serde(rename_all = "camelCase")]
#[ts(rename_all = "camelCase")]
System { file: AbsolutePathBuf },
/// Session-layer overrides supplied via `-c`/`--config`.
SessionFlags,
/// User config layer from a file (usually `config.toml`).
#[serde(rename_all = "camelCase")]
#[ts(rename_all = "camelCase")]
User { file: AbsolutePathBuf },
}
```
We want to rely on server-side auto-compaction instead of having the
client trigger context compaction manually. This API was stubbed as a
placeholder and never implemented.
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.
Our previous design of cancellation endpoint is not idempotent, which
caused a bunch of flaky tests. Make app server just returned a not_found
status instead of throwing an error if the login id is not found. Keep
V1 endpoint behavior the same.
This caused some conversations to not appear when they otherwise should.
Prior to this change, `thread/list`/`list_conversations_common` would:
- Fetch N conversations from `RolloutRecorder::list_conversations`
- Then it would filter those (like by the provided `model_providers`)
- This would make it potentially return less than N items.
With this change:
- `list_conversations_common` now continues fetching more conversations
from `RolloutRecorder::list_conversations` until it "fills up" the
`requested_page_size`.
- Ultimately this means that clients can rely on getting eg 20
conversations if they request 20 conversations.
- 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
