This PR makes an "insufficient quota" error fatal so we don't attempt to
retry it multiple times in the agent loop.
We have multiple bug reports from users about intermittent retry
behaviors, and this could explain some of them. With this change, we'll
eliminate the retries and surface a clear error message.
The PR is a nearly identical copy of [this
PR](https://github.com/openai/codex/pull/4837) contributed by
@abimaelmartell. The original PR has gone stale. Rather than wait for
the contributor to resolve merge conflicts, I wanted to get this change
in.
Currently, when the access token expires, we attempt to use the refresh
token to acquire a new access token. This works most of the time.
However, there are situations where the refresh token is expired,
exhausted (already used to perform a refresh), or revoked. In those
cases, the current logic treats the error as transient and attempts to
retry it repeatedly.
This PR changes the token refresh logic to differentiate between
permanent and transient errors. It also changes callers to treat the
permanent errors as fatal rather than retrying them. And it provides
better error messages to users so they understand how to address the
problem. These error messages should also help us further understand why
we're seeing examples of refresh token exhaustion.
Here is the error message in the CLI. The same text appears within the
extension.
<img width="863" height="38" alt="image"
src="https://github.com/user-attachments/assets/7ffc0d08-ebf0-4900-b9a9-265064202f4f"
/>
I also correct the spelling of "Re-connecting", which shouldn't have a
hyphen in it.
Testing: I manually tested these code paths by adding temporary code to
programmatically cause my refresh token to be exhausted (by calling the
token refresh endpoint in a tight loop more than 50 times). I then
simulated an access token expiration, which caused the token refresh
logic to be invoked. I confirmed that the updated logic properly handled
the error condition.
Note: We earlier discussed the idea of forcefully logging out the user
at the point where token refresh failed. I made several attempts to do
this, and all of them resulted in a bad UX. It's important to surface
this error to users in a way that explains the problem and tells them
that they need to log in again. We also previously discussed deleting
the auth.json file when this condition is detected. That also creates
problems because it effectively changes the auth status from logged in
to logged out, and this causes odd failures and inconsistent UX. I think
it's therefore better not to delete auth.json in this case. If the user
closes the CLI or VSCE and starts it again, we properly detect that the
access token is expired and the refresh token is "dead", and we force
the user to go through the login flow at that time.
This should address aspects of #6191, #5679, and #5505
## Summary
Duplicates the tests in `apply_patch_cli.rs`, but tests the freeform
apply_patch tool as opposed to the function call path. The good news is
that all the tests pass with zero logical tests, with the exception of
the heredoc, which doesn't really make sense in the freeform tool
context anyway.
@jif-oai since you wrote the original tests in #5557, I'd love your
opinion on the right way to DRY these test cases between the two. Happy
to set up a more sophisticated harness, but didn't want to go down the
rabbit hole until we agreed on the right pattern
## Testing
- [x] These are tests
In this PR, I am exploring migrating task kind to an invocation of
Codex. The main reason would be getting rid off multiple
`ConversationHistory` state and streamlining our context/history
management.
This approach depends on opening a channel between the sub-codex and
codex. This channel is responsible for forwarding `interactive`
(`approvals`) and `non-interactive` events. The `task` is responsible
for handling those events.
This opens the door for implementing `codex as a tool`, replacing
`compact` and `review`, and potentially subagents.
One consideration is this code is very similar to `app-server` specially
in the approval part. If in the future we wanted an interactive
`sub-codex` we should consider using `codex-mcp`
feature: Add "!cmd" user shell execution
This change lets users run local shell commands directly from the TUI by
prefixing their input with ! (e.g. !ls). Output is truncated to keep the
exec cell usable, and Ctrl-C cleanly
interrupts long-running commands (e.g. !sleep 10000).
**Summary of changes**
- Route Op::RunUserShellCommand through a dedicated UserShellCommandTask
(core/src/tasks/user_shell.rs), keeping the task logic out of codex.rs.
- Reuse the existing tool router: the task constructs a ToolCall for the
local_shell tool and relies on ShellHandler, so no manual MCP tool
lookup is required.
- Emit exec lifecycle events (ExecCommandBegin/ExecCommandEnd) so the
TUI can show command metadata, live output, and exit status.
**End-to-end flow**
**TUI handling**
1. ChatWidget::submit_user_message (TUI) intercepts messages starting
with !.
2. Non-empty commands dispatch Op::RunUserShellCommand { command };
empty commands surface a help hint.
3. No UserInput items are created, so nothing is enqueued for the model.
**Core submission loop**
4. The submission loop routes the op to handlers::run_user_shell_command
(core/src/codex.rs).
5. A fresh TurnContext is created and Session::spawn_user_shell_command
enqueues UserShellCommandTask.
**Task execution**
6. UserShellCommandTask::run emits TaskStartedEvent, formats the
command, and prepares a ToolCall targeting local_shell.
7. ToolCallRuntime::handle_tool_call dispatches to ShellHandler.
**Shell tool runtime**
8. ShellHandler::run_exec_like launches the process via the unified exec
runtime, honoring sandbox and shell policies, and emits
ExecCommandBegin/End.
9. Stdout/stderr are captured for the UI, but the task does not turn the
resulting ToolOutput into a model response.
**Completion**
10. After ExecCommandEnd, the task finishes without an assistant
message; the session marks it complete and the exec cell displays the
final output.
**Conversation context**
- The command and its output never enter the conversation history or the
model prompt; the flow is local-only.
- Only exec/task events are emitted for UI rendering.
**Demo video**
https://github.com/user-attachments/assets/fcd114b0-4304-4448-a367-a04c43e0b996
move the truncation logic to conversation history to use on any tool
output. This will help us in avoiding edge cases while truncating the
tool calls and mcp calls.
Adds a new ItemStarted event and delivers UserMessage as the first item
type (more to come).
Renames `InputItem` to `UserInput` considering we're using the `Item`
suffix for actual items.
## Summary
This PR is an alternative approach to #4711, but instead of changing our
storage, parses out shell calls in the client and reserializes them on
the fly before we send them out as part of the request.
What this changes:
1. Adds additional serialization logic when the
ApplyPatchToolType::Freeform is in use.
2. Adds a --custom-apply-patch flag to enable this setting on a
session-by-session basis.
This change is delicate, but is not meant to be permanent. It is meant
to be the first step in a migration:
1. (This PR) Add in-flight serialization with config
2. Update model_family default
3. Update serialization logic to store turn outputs in a structured
format, with logic to serialize based on model_family setting.
4. Remove this rewrite in-flight logic.
## Test Plan
- [x] Additional unit tests added
- [x] Integration tests added
- [x] Tested locally
# Tool System Refactor
- Centralizes tool definitions and execution in `core/src/tools/*`:
specs (`spec.rs`), handlers (`handlers/*`), router (`router.rs`),
registry/dispatch (`registry.rs`), and shared context (`context.rs`).
One registry now builds the model-visible tool list and binds handlers.
- Router converts model responses to tool calls; Registry dispatches
with consistent telemetry via `codex-rs/otel` and unified error
handling. Function, Local Shell, MCP, and experimental `unified_exec`
all flow through this path; legacy shell aliases still work.
- Rationale: reduce per‑tool boilerplate, keep spec/handler in sync, and
make adding tools predictable and testable.
Example: `read_file`
- Spec: `core/src/tools/spec.rs` (see `create_read_file_tool`,
registered by `build_specs`).
- Handler: `core/src/tools/handlers/read_file.rs` (absolute `file_path`,
1‑indexed `offset`, `limit`, `L#: ` prefixes, safe truncation).
- E2E test: `core/tests/suite/read_file.rs` validates the tool returns
the requested lines.
## Next steps:
- Decompose `handle_container_exec_with_params`
- Add parallel tool calls
### Title
## otel
Codex can emit [OpenTelemetry](https://opentelemetry.io/) **log events**
that
describe each run: outbound API requests, streamed responses, user
input,
tool-approval decisions, and the result of every tool invocation. Export
is
**disabled by default** so local runs remain self-contained. Opt in by
adding an
`[otel]` table and choosing an exporter.
```toml
[otel]
environment = "staging" # defaults to "dev"
exporter = "none" # defaults to "none"; set to otlp-http or otlp-grpc to send events
log_user_prompt = false # defaults to false; redact prompt text unless explicitly enabled
```
Codex tags every exported event with `service.name = "codex-cli"`, the
CLI
version, and an `env` attribute so downstream collectors can distinguish
dev/staging/prod traffic. Only telemetry produced inside the
`codex_otel`
crate—the events listed below—is forwarded to the exporter.
### Event catalog
Every event shares a common set of metadata fields: `event.timestamp`,
`conversation.id`, `app.version`, `auth_mode` (when available),
`user.account_id` (when available), `terminal.type`, `model`, and
`slug`.
With OTEL enabled Codex emits the following event types (in addition to
the
metadata above):
- `codex.api_request`
- `cf_ray` (optional)
- `attempt`
- `duration_ms`
- `http.response.status_code` (optional)
- `error.message` (failures)
- `codex.sse_event`
- `event.kind`
- `duration_ms`
- `error.message` (failures)
- `input_token_count` (completion only)
- `output_token_count` (completion only)
- `cached_token_count` (completion only, optional)
- `reasoning_token_count` (completion only, optional)
- `tool_token_count` (completion only)
- `codex.user_prompt`
- `prompt_length`
- `prompt` (redacted unless `log_user_prompt = true`)
- `codex.tool_decision`
- `tool_name`
- `call_id`
- `decision` (`approved`, `approved_for_session`, `denied`, or `abort`)
- `source` (`config` or `user`)
- `codex.tool_result`
- `tool_name`
- `call_id`
- `arguments`
- `duration_ms` (execution time for the tool)
- `success` (`"true"` or `"false"`)
- `output`
### Choosing an exporter
Set `otel.exporter` to control where events go:
- `none` – leaves instrumentation active but skips exporting. This is
the
default.
- `otlp-http` – posts OTLP log records to an OTLP/HTTP collector.
Specify the
endpoint, protocol, and headers your collector expects:
```toml
[otel]
exporter = { otlp-http = {
endpoint = "https://otel.example.com/v1/logs",
protocol = "binary",
headers = { "x-otlp-api-key" = "${OTLP_TOKEN}" }
}}
```
- `otlp-grpc` – streams OTLP log records over gRPC. Provide the endpoint
and any
metadata headers:
```toml
[otel]
exporter = { otlp-grpc = {
endpoint = "https://otel.example.com:4317",
headers = { "x-otlp-meta" = "abc123" }
}}
```
If the exporter is `none` nothing is written anywhere; otherwise you
must run or point to your
own collector. All exporters run on a background batch worker that is
flushed on
shutdown.
If you build Codex from source the OTEL crate is still behind an `otel`
feature
flag; the official prebuilt binaries ship with the feature enabled. When
the
feature is disabled the telemetry hooks become no-ops so the CLI
continues to
function without the extra dependencies.
---------
Co-authored-by: Anton Panasenko <apanasenko@openai.com>
The [official Rust
SDK](57fc428c57)
has come a long way since we first started our mcp client implementation
5 months ago and, today, it is much more complete than our own
stdio-only implementation.
This PR introduces a new config flag `experimental_use_rmcp_client`
which will use a new mcp client powered by the sdk instead of our own.
To keep this PR simple, I've only implemented the same stdio MCP
functionality that we had but will expand on it with future PRs.
---------
Co-authored-by: pakrym-oai <pakrym@openai.com>
Extracting tasks in a module and start abstraction behind a Trait (more
to come on this but each task will be tackled in a dedicated PR)
The goal was to drop the ActiveTask and to have a (potentially) set of
tasks during each turn
Adding the ability to resume conversations.
we have one verb `resume`.
Behavior:
`tui`:
`codex resume`: opens session picker
`codex resume --last`: continue last message
`codex resume <session id>`: continue conversation with `session id`
`exec`:
`codex resume --last`: continue last conversation
`codex resume <session id>`: continue conversation with `session id`
Implementation:
- I added a function to find the path in `~/.codex/sessions/` with a
`UUID`. This is helpful in resuming with session id.
- Added the above mentioned flags
- Added lots of testing
We need to construct the history different when compact happens. For
this, we need to just consider the history after compact and convert
compact to a response item.
This needs to change and use `build_compact_history` when this #3446 is
merged.
## 📝 Review Mode -- Core
This PR introduces the Core implementation for Review mode:
- New op `Op::Review { prompt: String }:` spawns a child review task
with isolated context, a review‑specific system prompt, and a
`Config.review_model`.
- `EnteredReviewMode`: emitted when the child review session starts.
Every event from this point onwards reflects the review session.
- `ExitedReviewMode(Option<ReviewOutputEvent>)`: emitted when the review
finishes or is interrupted, with optional structured findings:
```json
{
"findings": [
{
"title": "<≤ 80 chars, imperative>",
"body": "<valid Markdown explaining *why* this is a problem; cite files/lines/functions>",
"confidence_score": <float 0.0-1.0>,
"priority": <int 0-3>,
"code_location": {
"absolute_file_path": "<file path>",
"line_range": {"start": <int>, "end": <int>}
}
}
],
"overall_correctness": "patch is correct" | "patch is incorrect",
"overall_explanation": "<1-3 sentence explanation justifying the overall_correctness verdict>",
"overall_confidence_score": <float 0.0-1.0>
}
```
## Questions
### Why separate out its own message history?
We want the review thread to match the training of our review models as
much as possible -- that means using a custom prompt, removing user
instructions, and starting a clean chat history.
We also want to make sure the review thread doesn't leak into the parent
thread.
### Why do this as a mode, vs. sub-agents?
1. We want review to be a synchronous task, so it's fine for now to do a
bespoke implementation.
2. We're still unclear about the final structure for sub-agents. We'd
prefer to land this quickly and then refactor into sub-agents without
rushing that implementation.
We have two ways of loading conversation with a previous history. Fork
conversation and the experimental resume that we had before. In this PR,
I am unifying their code path. The path is getting the history items and
recording them in a brand new conversation. This PR also constraint the
rollout recorder responsibilities to be only recording to the disk and
loading from the disk.
The PR also fixes a current bug when we have two forking in a row:
History 1:
<Environment Context>
UserMessage_1
UserMessage_2
UserMessage_3
**Fork with n = 1 (only remove one element)**
History 2:
<Environment Context>
UserMessage_1
UserMessage_2
<Environment Context>
**Fork with n = 1 (only remove one element)**
History 2:
<Environment Context>
UserMessage_1
UserMessage_2
**<Environment Context>**
This shouldn't happen but because we were appending the `<Environment
Context>` after each spawning and it's considered as _user message_.
Now, we don't add this message if restoring and old conversation.
this dramatically improves time to run `cargo test -p codex-core` (~25x
speedup).
before:
```
cargo test -p codex-core 35.96s user 68.63s system 19% cpu 8:49.80 total
```
after:
```
cargo test -p codex-core 5.51s user 8.16s system 63% cpu 21.407 total
```
both tests measured "hot", i.e. on a 2nd run with no filesystem changes,
to exclude compile times.
approach inspired by [Delete Cargo Integration
Tests](https://matklad.github.io/2021/02/27/delete-cargo-integration-tests.html),
we move all test cases in tests/ into a single suite in order to have a
single binary, as there is significant overhead for each test binary
executed, and because test execution is only parallelized with a single
binary.
