## Description
This makes Codex Apps tool reads use a shared in-memory snapshot instead
of rereading the disk cache every time `list_all_tools()` runs. Disk
still seeds the cache on startup and gets updated after successful
fetches, but it is no longer the live read path.
The core change is that `McpManager` now owns a process-scoped
`CodexAppsToolsCache`. Codex threads in the same app-server process now
share this Codex Apps in-memory tools snapshot. The snapshot is keyed by
the Codex home plus the Codex Apps identity: the active Codex auth
user/workspace and the effective Codex Apps MCP source config.
There's already code to hard-refresh the cache, so we respect it in this
PR.
## Local benchmark
I ran a local steady-state microbenchmark of the exact repeated Codex
Apps cached-tools read this PR removes, using the same real local cache
payload in both trees: `3,678,138` bytes and `381` tools. The cache file
was already warm in the OS page cache, so this measures same-process
reread/deserialization work rather than cold-disk latency or full turn
latency. Each run is 25 iterations (mimicking a turn that makes 25
inference calls).
| Version | Run 1 | Run 2 | Avg |
|---|---:|---:|---:|
| `origin/main` disk read + JSON deserialize + `filter_tools` | `50.755
ms` | `52.894 ms` | `51.825 ms` |
| This branch in-memory `current_tools` + `filter_tools` | `0.740 ms` |
`0.778 ms` | `0.759 ms` |
That removes about `51 ms` from each repeated Codex Apps cached-tools
read on this machine, roughly `68x` faster for that subpath. It is
useful evidence for the hot path this PR changes, but not a claim that
every production turn gets `51 ms` faster; end-to-end impact also
depends on the rest of `list_all_tools()` and tool-payload construction.
This is on my M2 Max macbook, so with a slower disk this would be much
worse (and indeed we did see this really blew up turn runtime with a
slow disk).
## Summary
- track user-like input and tool-output boundaries in current-time
reminder state
- gate reminder injection when delivery_mode is
after_user_or_tool_output
- preserve interval debounce and forced reminders after context-window
changes
## Why
Training can request reminders only after user or tool-output items
while keeping the existing canonical pre-inference history-injection
path.
## Validation
- just test -p codex-core
current_time_reminders_can_follow_only_user_or_tool_outputs
- just test -p codex-core
current_time_reminders_follow_time_interval_and_persist_in_history
- just test -p codex-core
current_time_reminder_is_refreshed_after_compaction
- just fix -p codex-core
## tl;dr
Inject a `CODEX_PERMISSION_PROFILE` environment variable with the name
of the current permission profile when invoking a shell tool.
## Why
Shell tool owners may need to launch nested commands under the same
named permission profile, including through `codex sandbox -P PROFILE
--include-managed-config`. Until now, child processes could observe
sandbox and network metadata but could not identify the active named
permission profile.
The `--include-managed-config` flag is essential when a helper
reconstructs the sandbox from a profile name: it ensures the nested
sandbox also loads managed enterprise requirements. Without it, using
the inherited profile could unintentionally create a sandbox that does
not enforce the organization's managed restrictions.
The new environment value is intentionally informational and **must not
be treated as trusted input**. Any process in the ancestry can overwrite
an environment variable, so a consumer that passes this value to `codex
sandbox -P` must first validate it against the profiles that helper is
authorized to use.
## Example Use Case
Suppose an organization provides a trusted `remote-bash` wrapper that
lets Codex run a command on an approved build host. The local shell
command uses the named `:workspace` permission profile:
```toml
default_permissions = ":workspace"
```
The command exposed to the model is a small zsh wrapper. It deliberately
delegates with `exec`, preserving the original arguments and process
environment:
```zsh
#!/usr/bin/env zsh
exec /opt/codex-tools/remote_bash.py "$@"
```
The model invokes the public wrapper, not its Python implementation:
```sh
/opt/codex-tools/remote-bash \
--host builder.example.com \
-- printf '%s' 'hello world'
```
Only the inner implementation is authorized to escape the local sandbox:
```starlark
prefix_rule(
pattern=["/opt/codex-tools/remote_bash.py"],
decision="allow",
)
```
With zsh-fork, execution begins with `remote-bash` inside the
`:workspace` sandbox. When the wrapper calls `exec`, the exact prefix
rule matches `remote_bash.py`, so that inner script is restarted
unsandboxed. The escalated process inherits:
```text
CODEX_PERMISSION_PROFILE=:workspace
```
Inheritance does not make the value trustworthy. `remote_bash.py`
independently allowlists both the remote host and the permission profile
before using either value. In particular, a forged value such as
`:danger-full-access` is rejected before it can reach `codex sandbox
-P`:
```python
import argparse
import os
import shlex
import sys
ALLOWED_HOSTS = {"builder.example.com"}
ALLOWED_PROFILES = {":workspace"}
parser = argparse.ArgumentParser()
parser.add_argument("--host", required=True)
separator = sys.argv.index("--")
args = parser.parse_args(sys.argv[1:separator])
command = sys.argv[separator + 1:]
if args.host not in ALLOWED_HOSTS:
parser.error("host is not allowlisted")
if not command:
parser.error("the remote command must not be empty")
profile = os.environ.get("CODEX_PERMISSION_PROFILE")
if not profile:
raise SystemExit("CODEX_PERMISSION_PROFILE must not be empty")
if profile not in ALLOWED_PROFILES:
raise SystemExit("CODEX_PERMISSION_PROFILE is not allowlisted")
remote_command = shlex.join(command)
sandbox_command = shlex.join([
"codex", "sandbox", "-P", profile,
"--include-managed-config", "--",
"bash", "-lc", remote_command,
])
print(shlex.join(["ssh", args.host, sandbox_command]))
```
This builds each command layer as an argument vector and uses
`shlex.join()` at the boundary, rather than interpolating untrusted
shell text. After validation and parsing, the nested command has this
structure:
```text
ssh argv:
["ssh", "builder.example.com", SANDBOX_COMMAND]
SANDBOX_COMMAND argv:
["codex", "sandbox", "-P", ":workspace",
"--include-managed-config", "--",
"bash", "-lc", "printf %s 'hello world'"]
bash -lc payload argv:
["printf", "%s", "hello world"]
```
A production implementation could execute that SSH command. The
integration fixture prints it and parses the result back into arguments,
verifying the complete flow:
```text
model invokes outer wrapper
-> zsh-fork starts wrapper under :workspace
-> wrapper execs allowlisted Python script
-> prefix rule restarts Python script unsandboxed
-> Python script inherits CODEX_PERMISSION_PROFILE=:workspace
-> Python script verifies :workspace is allowlisted
-> remote command runs codex sandbox -P :workspace
with --include-managed-config
-> nested sandbox honors managed enterprise requirements
```
This gives the trusted helper access to resources outside the local
sandbox—such as SSH credentials—while ensuring that it can select only
an explicitly authorized profile and that work on the remote host
remains subject to the organization's managed requirements.
## What changed
- Inject `CODEX_PERMISSION_PROFILE` after shell environment policy
evaluation so the active profile wins over inherited or configured stale
values.
- Apply the variable to both `shell_command` and unified `exec_command`,
including local, zsh-fork, and remote exec-server paths.
- Remove stale values when the session has no active named profile.
- Preserve the current profile value when loading a shell snapshot so a
parent snapshot cannot restore an older profile.
## Testing
- Added classic-shell integration coverage proving an exact prefix rule
can run a `require_escalated` script outside the `:workspace` sandbox
while preserving `CODEX_PERMISSION_PROFILE=:workspace`.
- Added zsh-fork integration coverage in which the model invokes an
outer zsh wrapper, an inner allowlisted `remote_bash.py` runs
unsandboxed, and its printed SSH command reconstructs the inherited
`:workspace` sandbox with `--include-managed-config` while preserving
every argument after `--`.
- The example helper treats `CODEX_PERMISSION_PROFILE` as untrusted and
validates it against `ALLOWED_PROFILES` before constructing the nested
command.
- Assert that the reconstructed sandbox command includes
`--include-managed-config` so nested use of the inherited profile cannot
bypass managed enterprise requirements.
- Added coverage for overriding and removing stale profile values.
- Verified `shell_command` receives the selected active profile.
- Added shell snapshot coverage using `printenv
CODEX_PERMISSION_PROFILE`.
A zero interval lets callers request a reminder at every
otherwise-eligible inference boundary.
## Validation
- just test -p codex-core load_config_resolves_current_time_reminder
## Why
Helper threads such as task title generation can request a model ID that
is valid for the default OpenAI provider but unavailable from the active
provider. With Amazon Bedrock, `gpt-5.4-mini` is rejected while the
provider static catalog exposes Bedrock model IDs such as
`openai.gpt-5.5` and `openai.gpt-5.4`. This causes repeated background
404s and can surface a misleading turn error even when the main turn
succeeds.
Clients need an explicit way to ask app-server to resolve an unavailable
helper model to the active provider default. That fallback must remain
limited to providers with an authoritative static catalog so custom or
dynamically discovered model IDs are not rewritten based on an
incomplete catalog.
Fixes#28741.
## What changed
- Add the experimental `allowProviderModelFallback` option to
`thread/start`, defaulting to `false` to preserve existing behavior.
- Thread the option through thread creation and model selection.
- When enabled for a static model manager, preserve requested models
present in the catalog and replace unavailable models with the provider
default.
- Continue preserving explicit model IDs for dynamic model managers
without fetching a catalog solely to validate them.
- Document the new `thread/start` behavior in the app-server API
overview.
## Test
Temporary test-client harness:
```
ThreadStartParams {
model: Some("gpt-5.4-mini".to_string()),
allow_provider_model_fallback: true,
..Default::default()
}
```
Command:
```
CODEX_HOME=/tmp/codex-bedrock-thread-start-home \
CODEX_E2E_BEDROCK_THREAD_START_ONLY=1 \
./target/debug/codex-app-server-test-client \
--codex-bin ./target/debug/codex \
-c 'model_provider="amazon-bedrock"' \
send-message-v2 --experimental-api ignored
```
Relevant output:
```
> "method": "thread/start",
> "params": {
> "model": "gpt-5.4-mini",
> "modelProvider": null,
> "allowProviderModelFallback": true,
> ...
> }
< "result": {
< "model": "openai.gpt-5.5",
< "modelProvider": "amazon-bedrock",
< ...
< }
```
## Why
`selectedCapabilityRoots` is durable thread intent: “use this capability
root from environment `worker`.”
The important product assumption is:
> One environment ID always names the same logical executor and stable
contents.
`worker` does not silently change from executor A to an unrelated
executor B. The process-local connection handle for `worker` can still
be replaced while Codex is running, though, for example when
`environment/add` registers a fresh handle for the same logical
environment.
The thread should persist only the stable selection. Each model step
should pair that selection with the exact ready handle captured for that
step.
## The boundary
```text
persisted thread intent
plugin@1 -> environment "worker"
|
| capture the current step
v
model-step view
unavailable, or
plugin@1 + worker's exact captured ready handle
```
The environment ID is the stable identity and cache key. The
`Arc<Environment>` is only a process-local handle retained so consumers
of one model step use the same captured environment. It is never
persisted and it does not imply different environment contents.
## What changes
### Persist the stable selection
Selected roots are written into `SessionMeta` and restored with the
thread. Forked subagents inherit the same selections, including
bounded-history forks.
Only stable data is persisted: root ID, environment ID, and root path.
### Capture readiness together with the exact handle
The environment snapshot records:
```rust
environment_id -> Some(Arc<Environment>) // ready in this step
environment_id -> None // still starting in this step
```
This prevents readiness and execution from coming from different
registry snapshots.
For example:
```text
step snapshot: worker -> handle A, ready
environment/add: worker -> fresh handle B for the same logical environment
current step: plugin@1 still uses captured handle A
```
Without carrying handle A in the snapshot, the resolver could combine “A
was ready” with handle B and treat B as ready before it had finished
starting.
This does not change cache invalidation. Stable capability metadata
remains identified by environment ID and capability root. Replacing a
process-local handle under the same stable environment ID does not
invalidate or rediscover that metadata.
### Resolve availability per model step
- A ready captured environment produces resolved roots using its
captured handle.
- A starting, missing, or failed environment is omitted from that step.
- A selected lazy environment that is outside the turn's captured
environment set is asked to start, and a later step can observe it as
ready.
- No capability files are scanned here.
Transient transport disconnects remain the remote client's reconnect
concern. This PR models initial attachment/readiness; it does not add
live socket-connectivity state.
## Example
```text
thread selection: plugin@1 -> environment "worker"
step 1: worker is starting -> plugin@1 unavailable
step 2: worker is ready -> plugin@1 resolves through worker's captured handle
step 3: fresh local handle -> current step remains pinned; a later step captures its own view
```
Temporary unavailability does not discard the durable selection. Later
PRs can retain stable metadata caches while projecting only currently
available capabilities into model-visible World State.
## Compatibility
The app-server request shape does not change. Older rollouts without
`selected_capability_roots` deserialize to an empty list.
## Stack
1. **This PR:** persist stable selected roots and resolve them through
an exact model-step handle.
2. #29960: cache stable skill metadata and project available skills into
World State.
3. #29946: cache stable plugin declarations and manage the separate live
MCP runtime.
## Why
#28522 routes selected-plugin HTTP MCP traffic through the owning
executor, but OAuth bootstrap and refresh still used host-local clients.
Executor-only servers therefore cannot complete discovery or login
through the same network boundary as the MCP connection.
## What changed
- adapt `codex_exec_server::HttpClient` to RMCP 1.8's `OAuthHttpClient`
contract
- let RMCP own discovery, dynamic registration, PKCE, token exchange,
and refresh
- route auth status, persisted-token startup, and app-server login
through the server runtime while preserving the existing local discovery
path
- add optional `threadId` to `mcpServer/oauth/login` and echo it in the
completion notification
- implement RMCP's redirect policy and 1 MiB OAuth response limit over
executor HTTP
- cover selected-thread OAuth discovery and login through an
executor-only route
Depends on #28522.
## Why
Older rollouts can retain model-visible context for a WorldState section
without having a persisted snapshot for that section. Treating the
missing snapshot as definitely absent can duplicate old context or fail
to tell the model that it was replaced or removed.
This provides a generic migration path for sections moving into
WorldState, beginning with AGENTS.md.
Builds on #29810.
## What changed
- distinguish section state that is absent, known from a persisted
snapshot, or unknown because matching legacy context remains in history
- let WorldState sections identify their own legacy fragments while
`ContextManager` owns history reconciliation and baseline persistence
- make AGENTS.md emit one conservative replacement or removal update for
legacy history, then deduplicate from the newly persisted baseline
- preserve existing environment rendering when persisted section data is
missing or malformed
## Testing
- `just test -p codex-core world_state`
- `just test -p codex-core
cold_resume_invalidates_deleted_legacy_agents_md_once -- --exact`
## Why
With deferred executors, a turn can begin before a remote environment
attaches. AGENTS.md discovery previously ran only during session setup,
so instructions from a later environment never reached the model or the
session instruction sources.
WorldState persistence has now landed, so this uses the durable
model-visible baseline directly instead of carrying a temporary
resume/fork compatibility path.
## What
- Add an `AgentsMdManager` in `SessionServices` to own host
instructions, loaded state, and refresh caching.
- When `DeferredExecutor` is enabled, refresh AGENTS.md when attached
environment selections change and freeze the result in the corresponding
`StepContext`.
- Represent AGENTS.md as a persisted WorldState section for every
session, with bounded initial, replacement, and removal updates.
- Remove duplicate AGENTS.md state and rendering from
`SessionConfiguration` and `TurnContext`.
- Build initial context, per-request updates, and compaction context
from the same step-scoped value.
- On resume and fork, compare current instructions with the restored
WorldState baseline and inject a replacement exactly once when they
differ.
Builds on #29833, #29835, and #29837.
## Tests
- Covers a remote environment becoming ready mid-turn, with AGENTS.md
appearing on the next request exactly once and updating canonical
instruction sources.
- Covers full, unchanged, replaced, and removed AGENTS.md WorldState
rendering.
- Covers changed instructions across cold resume and fork without
duplicate reinjection.
- Covers remote-v2 compaction retaining creation-time instructions in
the live session and cold resume appending one replacement when the
source changed.
- Ran focused `codex-core` AGENTS.md, WorldState, and context-update
test suites.
## Stack
This is PR 3 of the simplified HAI single-run-task stack:
- [#19047](https://github.com/openai/codex/pull/19047) Agent Identity
assertion and task-registration primitives, including the shared
run-task helper used by existing Agent Identity JWT auth.
- [#19049](https://github.com/openai/codex/pull/19049)
Disabled-by-default ChatGPT auth opt-in that provisions/reuses persisted
Agent Identity runtime auth and its single run task.
- [#19051](https://github.com/openai/codex/pull/19051) Run-scoped
provider auth that uses one backend-owned task id for first-party
inference and compaction requests.
[#19054](https://github.com/openai/codex/pull/19054) collapsed out of
the active stack because the simplified design no longer needs a
separate background/control-plane task helper.
## Summary
This PR moves Agent Identity usage into provider auth resolution. That
keeps `AgentAssertion` auth tied to first-party OpenAI provider requests
instead of applying a late session-wide override that could affect
local, custom, Bedrock, API-key, or external-bearer providers.
What changed:
- adds a small `ProviderAuthScope` struct carrying the run auth policy
and session source needed by provider-scoped auth resolution
- lets `Session` opt the existing `ModelClient` into `ChatGptAuth`
policy when `use_agent_identity` is enabled, without adding a second
model-client constructor
- resolves Agent Identity only for first-party OpenAI provider auth
paths
- uses the persisted run task id from the `AgentIdentityAuth` record to
build `AgentAssertion` auth for Responses requests
- routes shared request setup through scoped provider auth so unary
compact requests use the same run-task assertion path as inference turns
- keeps local/custom/Bedrock/env-key/external-bearer provider auth
unchanged
- lets missing run-task state surface through the existing model-request
error path instead of silently falling back to bearer auth
This PR intentionally does not create thread-scoped, target-scoped, or
background-scoped task identities. The run task is the only task Codex
registers in this POC shape.
## Testing
- `just test -p codex-model-provider`
- `just test -p codex-core client::tests::provider_auth_scope_uses`
- `just test -p codex-core remote_compact_uses_agent_identity_assertion`
## Summary
- add a cancellable sleep operation to `TimeProvider`
- route `clock.sleep` through the configured provider
- extend the supported sleep duration to 12 hours
- complete the sleep turn item before propagating provider failures
## Why
This isolates the core clock abstraction needed by external clock
integrations. Existing system and app-server behavior remains wall-clock
based in this PR; the stacked follow-up supplies app-server sleeps from
an external clock.
## Why
Token-budget reminder and guidance messages can require more than 1,000
bytes to provide useful model-facing instructions. At the same time,
these strings are injected into model-visible context, so their size
must remain tightly bounded in response to the P0 context-growth
concern. A 2,000-byte runtime cap provides additional room without
allowing the substantially larger context growth of a 4 KiB limit.
## What changed
- raises the runtime byte limits for token-budget reminder templates and
guidance messages from 1,000 to 2,000
- raises the corresponding JSON Schema `maxLength` values to 2,000
- regenerates `codex-rs/core/config.schema.json`
## Testing
- `just test -p codex-features`
- `just test -p codex-core load_config_resolves_token_budget_config
load_config_rejects_invalid_token_budget_reminder_template`
The full `codex-core` test run completed 2,858 tests successfully and
encountered seven unrelated environment-sensitive failures involving
Seatbelt/network environment assertions, MCP capability setup, and abort
timing.
## Why
MCP error-code telemetry special-cased Codex Apps: its reported error
codes were retained, while codes from every other MCP server were
replaced with `unknown`. Error reporting should behave consistently for
every MCP server. The server name already identifies where an error came
from, so telemetry does not need a separate Codex Apps classification.
This follows up on [#28976](https://github.com/openai/codex/pull/28976),
which introduced MCP error-code telemetry.
## What changed
- Add the MCP server name to call, duration, and error metrics.
- Retain bounded, sanitized tool error codes from every MCP server.
- Remove `McpErrorCodeSource` and the Codex Apps ownership lookup from
telemetry collection.
- Use the same metric-tagging path for blocked, rejected, and executed
MCP calls.
## Test plan
- Verify the complete metric tag set includes the sanitized MCP server
name.
- Verify error codes from ordinary MCP servers are retained, bounded,
and sanitized.
- Preserve coverage for request failures, tool-result failures, nested
auth failures, and span attributes.
## Why
Persisting `WorldState` snapshots and patches is only useful if resume
and fork restore that exact comparison baseline. Rebuilding it from
`TurnContextItem` loses section state and can either repeat or suppress
model-visible updates.
This is the third PR in the WorldState persistence stack, built on
#29835.
## What
- Replay full WorldState snapshots and RFC 7386 patches through the
existing rollout reconstruction segments.
- Discard state from rolled-back turns and treat compaction as a
baseline reset.
- Hydrate `ContextManager` from the reconstructed snapshot on resume and
fork.
- Remove the synthetic `TurnContextItem` to WorldState conversion path.
- Leave legacy or malformed rollouts without a baseline so the next
update safely emits a full snapshot.
## Testing
- `just test -p codex-core world_state`
- `just test -p codex-core rollout_reconstruction_tests`
- `just fix -p codex-core`
- `just test -p codex-core` *(the changed tests passed; the full run
also hit unrelated existing/test-environment failures, primarily a
missing `test_stdio_server` binary)*
## Why
Ultra should be one user-facing reasoning selection for work that
benefits from both maximum reasoning and proactive multi-agent
delegation. Without it, clients must coordinate maximum reasoning with
the experimental `multiAgentMode` setting, even though the inference
backend still expects its existing `max` effort value.
This change makes reasoning effort the source of truth: clients select
`ultra`, core derives proactive multi-agent behavior when the turn is
eligible for multi-agent V2, and inference requests continue to use the
backend-compatible `max` value.
## What changed
- Add `ultra` as a first-class reasoning effort and preserve
model-catalog ordering when exposing it to clients.
- Convert `ultra` to `max` at the inference request boundary, including
Responses HTTP/WebSocket requests, startup prewarm, compaction, and
memory summarization.
- Derive effective multi-agent mode per turn from effective reasoning
effort:
- eligible multi-agent V2 + `ultra` → `proactive`
- eligible multi-agent V2 + any other effort → `explicitRequestOnly`
- V1 or otherwise ineligible sessions → no multi-agent mode instruction
- Keep the derived effective mode in turn context history so successive
turns can emit a developer-message update only when the effective mode
changes.
- Remove selected multi-agent mode from core session configuration, turn
construction, thread settings, resume/fork restoration, and subagent
spawn plumbing. Subagents inherit reasoning effort and derive their own
effective mode.
- Retain the experimental app-server `multiAgentMode` fields for wire
compatibility while marking them deprecated. Request values are accepted
but ignored; compatibility response fields report `explicitRequestOnly`.
- Display Ultra in the TUI using the order supplied by `model/list`.
## Validation
- `just test -p codex-core ultra_reasoning_uses_max_for_requests`
- `just test -p codex-tui model_reasoning_selection_popup`
## Why
`WorldState` currently remembers its model-visible diff baseline only in
memory. That leaves no durable source for restoring the exact baseline
after resume, fork, rollback, or compaction.
This is the second PR in the WorldState persistence stack, built on
#29833 and following #29249. It records durable state transitions; the
next PR will replay them during rollout reconstruction.
## What
- Add a `world_state` rollout item containing either a full snapshot or
an RFC 7386 JSON Merge Patch.
- Persist a full snapshot after initial context and after compaction
establishes a new context window.
- Persist non-empty patches when later sampling steps or turns advance
the WorldState baseline.
- Write model-visible history before its matching WorldState record, so
an interrupted write can only cause a safe repeated update on replay.
- Preserve WorldState records for full-history forks while excluding
them from thread previews, metadata, and app-server history
materialization.
Older binaries read rollout lines independently, so they skip the
unknown `world_state` records while retaining the rest of the thread.
## Testing
- `just test -p codex-core
snapshot_merge_patch_changes_and_removes_nested_values`
- `just test -p codex-core
world_state_baseline_deduplicates_until_history_is_replaced`
- `just test -p codex-core
deferred_executor_compaction_preserves_then_updates_environment_once`
- `just test -p codex-protocol`
- `just test -p codex-rollout`
- `just test -p codex-state`
- `just test -p codex-thread-store`
- `just test -p codex-app-server-protocol`
## Why
MCP authentication has distinct OAuth and ChatGPT-session flows.
Representing that choice as `use_chatgpt_auth` makes one flow implicit
and allows the configuration model to express the distinction only
through a boolean.
ChatGPT credential forwarding also needs a first-party trust boundary. A
configurable `chatgpt_base_url` controls routing, but must not grant an
MCP server permission to receive session credentials.
This change builds on #29733, where the boolean was introduced.
## What changed
- Replace `use_chatgpt_auth` with an `auth` field backed by the
exhaustive `McpServerAuth` enum.
- Support `auth = "oauth"` and `auth = "chatgpt"`, with OAuth remaining
the default.
- Trust only the origin derived from the existing hardcoded
`CHATGPT_CODEX_BASE_URL` when granting ChatGPT auth to an MCP server.
- Keep configured bearer tokens and authorization headers ahead of the
selected authentication flow.
- Update config writers, schema output, fixtures, and integration-test
setup to use the enum.
## Verification
Integration coverage exercises the complete streamable HTTP startup path
in two independent configurations:
- A directly constructed MCP configuration verifies that matching an
overridden `chatgpt_base_url` does not grant ChatGPT auth.
- A persisted `config.toml` containing an attacker-controlled
`chatgpt_base_url` and `auth = "chatgpt"` verifies the same boundary
through normal config parsing.
Both tests complete MCP initialization and tool listing and assert that
the full captured request sequence contains no authorization headers.
Separate integration coverage verifies that configured authorization
takes precedence over ChatGPT auth.
## Why
`WorldState` currently keeps its diff baseline as live Rust objects
keyed by process-local `TypeId`. That baseline cannot be written to a
rollout or restored after resume, so Codex reconstructs an approximation
from `TurnContextItem`.
This is the first change in the WorldState persistence stack. It gives
every section a stable persisted identity and a compact serializable
comparison snapshot without changing rollout behavior yet.
## What changed
- Require each `WorldStateSection` to define a stable ID and
serializable snapshot type.
- Reject duplicate section IDs when constructing `WorldState`.
- Persist a dedicated environment comparison snapshot using
model-visible strings instead of runtime path types.
- Store only `WorldStateSnapshot` in `ContextManager`, removing the
parallel live-object baseline.
- Render diffs by restoring each section's typed snapshot; invalid
snapshots fall back to a full section render.
- Omit null object fields for future RFC 7386 patches while preserving
null values inside arrays.
Follow-up PRs will record full snapshots and merge patches, then restore
the baseline during resume, fork, and rollback.
## Test plan
- WorldState snapshot tests cover stable IDs, duplicate rejection, null
omission, and array preservation.
- Environment tests cover persistence-safe snapshot values and existing
diff rendering.
- ContextManager baseline deduplication and session context-update
persistence tests.
Related: #29249
## Why
ChatGPT session authentication was inferred from the reserved Codex Apps
server name. That couples credential routing to Codex Apps-specific
behavior and prevents other MCP endpoints hosted by ChatGPT from
explicitly using the current session.
The opt-in also needs a clear security boundary: an arbitrary MCP
configuration must not be able to redirect ChatGPT credentials to
another origin.
## What changed
- Add `use_chatgpt_auth` to HTTP MCP server configuration, defaulting to
`false`.
- Honor the setting only when the parsed server URL has the same HTTP(S)
origin as the configured `chatgpt_base_url`; otherwise remove the
capability before startup.
- Resolve bearer tokens and static or environment-backed authorization
headers before selecting authentication, with configured authorization
taking precedence over ChatGPT session auth.
- Enable the setting for the built-in Codex Apps and hosted plugin
runtime endpoints while keeping Codex Apps caching and tool
normalization scoped to the reserved server.
- Persist the setting through MCP config rewrite paths and expose it in
the generated config schema.
- Load the current login state for `codex mcp list` so reported auth
status matches runtime behavior.
## Verification
Core integration coverage exercises the complete streamable HTTP MCP
startup path and verifies that:
- a same-origin opted-in server receives the current ChatGPT access
token;
- an explicitly configured authorization header takes precedence;
- a different-origin server completes MCP initialization and tool
listing without receiving any ChatGPT authorization header.
## Why
This PR adds a configurable `<context_window_guidance>` developer
section immediately after `<context_window>`. Harness integrations need
this section to give the model deployment-specific instructions for
preparing for context-window transitions.
## What changed
- Add an optional `features.token_budget.guidance_message` config with a
1,000-byte runtime cap and generated schema support.
- Render configured guidance as a developer `ContextualUserFragment`
wrapped in `<context_window_guidance>` immediately after
`<context_window>`.
- Omit the section when guidance is unset, empty, or whitespace-only.
- Preserve the resolved value in config locks and classify persisted
guidance as contextual developer content.
- Add integration coverage for rendered content and ordering.
## Summary
- move sleep tool enablement from top-level `[features].sleep_tool` to
`[features.current_time_reminder].sleep_tool`
- remove the standalone `Feature::SleepTool` flag and gate `clock.sleep`
from resolved current-time configuration
- update config schema, config-lock materialization, and existing sleep
coverage
Stacked on #29907.
## Summary
- expose the interruptible sleep tool as `clock.sleep` instead of
top-level `sleep`
- keep `clock.curr_time` and `clock.sleep` in the same model-visible
namespace when both features are enabled
- update existing core and app-server integration coverage to issue
namespaced sleep calls
## Why
Sleep is a clock operation. Grouping it with `clock.curr_time` gives the
model a more coherent tool surface without changing the sleep feature
gate or runtime behavior.
## Validation
- `just test -p codex-core sleep_tool_follows_feature_gate`
- `just test -p codex-core any_new_input_interrupts_sleep`
- `just test -p codex-app-server
sleep_emits_started_and_completed_items`
## Why
Connector declarations currently enter Codex through broad plugin
capability summaries, then MCP setup, turn tooling, and `app/list` each
reconstruct the same information. That makes executor-selected
connectors difficult to add without coupling connector behavior to the
host plugin loader.
This PR introduces a small connector-owned value that later stack layers
can populate before thread startup.
## What changed
- Move the pure app-declaration parser into `codex-connectors`,
preserving declaration order and category cleanup while leaving
host-side validation and deduplication unchanged.
- Add an immutable `ConnectorSnapshot` with ordered connector IDs and
plugin display-name provenance.
- Adapt the existing local-plugin capability summaries into that
snapshot at current consumer boundaries.
- Use the snapshot for MCP tool provenance, turn connector inventory,
and `app/list`.
- Keep the crate API narrow: no test-only snapshot accessors are
exposed.
The externally visible behavior is unchanged. Connector tools still come
from the orchestrator-owned `/ps/mcp` server, and local plugin
enablement remains owned by the existing plugin loader.
## Stack scope
This is the foundation only. It does not read selected executor packages
or change thread startup. #29852 adds the executor-backed declaration
reader, and #29856 composes selected declarations into a thread
snapshot.
## Why
When Codex uses a remote `ExecutorFileSystem`, every `get_metadata` call
is an exec-server round trip. Upward discovery currently pays those
round trips serially in two latency-sensitive places:
- session startup, while locating the configured project root before
loading `AGENTS.md`; and
- Git-root discovery, which runs before per-turn Git diff enrichment.
The goal is to remove the serial ancestor dependency without adding a
new filesystem RPC, JSON-RPC batch method, Git executable dependency, or
cache.
## Example
Assume this layout, with `.git` as the configured project-root marker:
```text
/workspace/repo/.git
/workspace/repo/AGENTS.md
/workspace/repo/crates/core/ <- cwd
```
The marker probes have this required precedence:
```text
1. /workspace/repo/crates/core/.git
2. /workspace/repo/crates/.git
3. /workspace/repo/.git
4. /workspace/.git
5. /.git
```
Previously, probe 2 was not sent until probe 1 returned, and probe 3 was
not sent until probe 2 returned. With this change, the client lazily
keeps up to eight ordinary `fs/getMetadata` requests in flight, but
consumes their results in the order above. Codex must still learn that
probes 1 and 2 are absent before accepting probe 3, so the nearest root
always wins. Once probe 3 succeeds, the client has its answer and stops
awaiting probes 4 and 5. Requests that were already sent may still
finish on the worker.
For the marker phase alone, with a 50 ms client-to-worker round trip and
fast local metadata calls, finding the root at probe 3 changes from
roughly three serialized round trips (150 ms) to one round trip plus
worker processing. The later `AGENTS.md` candidate phase remains
separate and ordered.
Only after `/workspace/repo` is selected does `AGENTS.md` discovery
check instruction candidates, in root-to-cwd order:
```text
/workspace/repo/AGENTS.override.md
/workspace/repo/AGENTS.md
/workspace/repo/crates/AGENTS.override.md
/workspace/repo/crates/AGENTS.md
/workspace/repo/crates/core/AGENTS.override.md
/workspace/repo/crates/core/AGENTS.md
```
The first configured candidate found in each directory wins. These
checks remain ordered and no instruction candidate above
`/workspace/repo` is issued. Git-root discovery uses the same bounded
lookup with only `.git` as the marker.
## What changed
- Added a client-side find-up helper that generates `ancestor x marker`
probes lazily, nearest directory first and configured marker order
within each directory.
- Uses an ordered concurrency window of eight scalar metadata requests.
This bounds executor load while preserving nearest-root and marker
precedence.
- Reuses the helper for both configured project-root discovery and
remote Git-root discovery.
- Keeps Git ancestor and marker construction in `AbsolutePathBuf`,
converting only each complete `.git` probe to `PathUri`. This preserves
native paths that require an opaque URI fallback, such as Windows
namespace paths.
- Preserves existing error behavior: `AGENTS.md` discovery propagates
non-`NotFound` metadata errors, while Git discovery treats a failed
marker probe as absent and continues upward.
- Reads each discovered `AGENTS.md` directly instead of statting it a
second time.
No filesystem trait or exec-server protocol method is added. An empty
`project_root_markers` list performs no ancestor-marker I/O and checks
instruction candidates only in `cwd`. This change also deliberately does
not cache roots across turns.
## Symlinks
Upward traversal remains **lexical**. The helper does not canonicalize
`cwd`; it appends marker names to the supplied path and walks that
path's textual parents. The filesystem performs the actual metadata/read
operation, and the current local and exec-server implementations follow
live symlink targets.
For example:
```text
/tmp/pkg -> /workspace/repo/packages/pkg
cwd = /tmp/pkg/src
actual Git marker = /workspace/repo/.git
```
The lexical probes are `/tmp/pkg/src/.git`, `/tmp/pkg/.git`,
`/tmp/.git`, and `/.git`. They do not jump from `/tmp/pkg` to the
target's parent `/workspace/repo`, so this spelling of `cwd` does not
discover `/workspace/repo/.git`. That is the existing behavior and is
unchanged by this PR.
Conversely, if `/tmp/repo -> /workspace/repo`, then probing
`/tmp/repo/.git` follows the directory symlink and finds
`/workspace/repo/.git`; the reported root remains the lexical path
`/tmp/repo`. A live symlink used directly as `.git`, another configured
marker, or `AGENTS.md` is also followed. A symlinked `AGENTS.md` is
loaded when its target is a regular file, while a broken symlink behaves
as `NotFound`.
Summary
- Emit `codex_plugin_install_requested` when a validated plugin install
request is made, before the user accepts or declines the elicitation.
- Record the exact model-visible plugin ID, remote plugin ID, required
connector IDs, stable suggestion ID, and `endpoint_recommendation` vs
`legacy_discovery` source.
- Keep `suggest_reason` out of telemetry and leave connector-only
install requests unchanged.
Rollout
- Backend/schema dependency:
https://github.com/openai/openai/pull/1065270
- Land the backend PR before this producer starts sending the event.
Validation
- `just test -p codex-analytics` (83 passed)
- `just test -p codex-core request_plugin_install` (17 passed)
- `just fix -p codex-analytics`
- `just fix -p codex-core`
- `just fmt`
- `git diff --check`
## Why
Core and tools need to request MCP elicitation without constructing
app-server wire payloads. The request should remain a neutral protocol
concept until app-server serializes it for a client.
## What changed
- Switched core and tools to
`codex_protocol::approvals::ElicitationRequest`.
- Derived turn and server context inside core instead of carrying
app-server request types through lower layers.
- Kept the app-server payload unchanged through an explicit boundary
conversion.
- Removed the remaining production app-server-protocol dependency from
tools.
## Stack
This is PR 5 of 6, stacked on [PR
#29723](https://github.com/openai/codex/pull/29723). Review only the
delta from `codex/split-connector-metadata-types`. Next: [PR
#29725](https://github.com/openai/codex/pull/29725).
## Validation
- `codex-core` MCP coverage passed: 87 tests.
- Tools elicitation and app-server round-trip coverage passed.
## Summary
- Add `iconAssets` and `iconDarkAssets` to the app-list protocol.
- Preserve structured icons through directory merging and the connector,
app-
server, and TUI boundaries.
- Keep legacy logo URLs unchanged as compatibility fallbacks.
- Update generated protocol schemas and TypeScript types.
Pick up the AgentGraphStore migration.
- Inject an explicit optional agent graph store into `ThreadManager`
- Move all calls to spawn, close, recursive resume, and
subtree/archive/delete/feedback traversal through it
- Keep using `LocalAgentGraphStore` when SQLite is available
This required some changes to the interface to deal with futures:
- The interface now matches `ThreadStore`'s object-safe pattern by
returning a boxed `AgentGraphStoreFuture` directly, allowing
`ThreadManager` to hold `Arc<dyn AgentGraphStore>`
*Slight behavior change!* Unfiltered subtree enumeration now performs a
single all-status breadth-first traversal, so a closed grandchild
beneath an open edge is included; the previous Open-then-Closed
traversals could not cross mixed-status paths and silently omitted it.
## Why
`thread/list` can filter direct children with `parentThreadId`, but
clients cannot request an entire spawned subtree. Discovering every
descendant requires repeated client-side requests and gives up the
database's existing filtering and pagination path.
## What changed
Experimental clients can use `ancestorThreadId` to return strict
descendants at any depth while `parentThreadId` retains its direct-child
meaning. The filters are mutually exclusive, the ancestor is excluded,
and every result preserves its immediate `parentThreadId` so callers can
reconstruct the tree.
## How it works
- **Explicit relationship:** Internal list parameters distinguish direct
children from transitive descendants without changing the meaning of
`parentThreadId`.
- **Existing graph:** Persisted parent-child spawn edges remain the
source of truth, so descendant lookup needs no schema migration or
ancestry cache.
- **Indexed traversal:** A recursive SQLite query starts from the
parent-edge index, walks each generation, and applies thread filters,
sorting, and cursor pagination in the same database request.
- **Reconstructable results:** The response stays flat and normally
ordered while carrying each descendant's immediate parent.
## Verification
Ran 550 tests across the protocol, state, rollout, and thread-store
crates, then reran the four focused state, store, and app-server
descendant-listing tests after the final diff reduction. Scoped Clippy
and formatting checks passed. Stable and experimental schema generation
was checked; the stable fixtures remain unchanged while the experimental
schema includes the new field.
## Why
Inter-agent messages are recorded in live history as
`ResponseItem::AgentMessage`, but rollouts stored
`InterAgentCommunication` and rebuilt the response item during resume.
This made the rollout differ from the actual Responses history.
## What changed
- store the prepared `agent_message` response item directly
- keep `trigger_turn` in a small local metadata record for fork
truncation
- keep reading older `inter_agent_communication` rollout items
## Why
Topology-neutral app-server integration tests should exercise automatic
environment selection so the same setup covers local and remote
executors.
## What
Migrate eligible tests to `TestAppServer::new_with_auto_env()` and
`send_thread_start_request_with_auto_env()`. Leave explicit-topology
tests unchanged, and skip the request-permissions case on Windows with a
TODO for cross-platform tool routing.
## Validation
- `just test -p codex-app-server`
- `bazel test //codex-rs/app-server:app-server-all-wine-exec-test
--test_output=errors`
Stacked on #29788.
## Why
Made a mistake when carving #29746 out of my local changes and the test
was missing from the build graph. Oops!
## What
Enable the app-server Wine exec test target. Remove the `manual` tag
from generated Wine-exec test variants so wildcard Bazel test
invocations select them. Refactor the smoke test to ensure it passes
with current Windows support.
## Why
Connector metadata is consumed by connector discovery, ChatGPT
integration, core, and TUI code. Treating app-server's wire DTO as the
shared domain model reverses the intended dependency direction.
## What changed
- Added connector-owned app branding, review, screenshot, metadata, and
info types.
- Added explicit conversions in app-server and TUI while preserving
app-server's wire payloads.
- Removed production app-server-protocol dependencies from connectors
and ChatGPT connector code.
## Stack
This is PR 4 of 6, stacked on [PR
#29722](https://github.com/openai/codex/pull/29722). Review only the
delta from `codex/split-config-layer-types`. Next: [PR
#29724](https://github.com/openai/codex/pull/29724).
## Validation
- Connector and tools coverage passed.
- App-server app-list coverage passed: 13 tests.
## Why
Config layer provenance describes how effective configuration was
assembled, so it belongs with the config loader rather than in
app-server's serialized API types.
## What changed
- Moved `ConfigLayerSource`, `ConfigLayerMetadata`, and `ConfigLayer`
ownership into `codex-config`.
- Kept app-server's wire payloads unchanged and added explicit
conversions at the app boundary.
- Removed lower-level app-server-protocol dependencies from config
consumers.
## Stack
This is PR 3 of 6, stacked on [PR
#29721](https://github.com/openai/codex/pull/29721). Review only the
delta from `codex/split-auth-domain-types`. Next: [PR
#29723](https://github.com/openai/codex/pull/29723).
## Validation
- `codex-config` coverage passed.
- App-server config-manager and config RPC coverage passed.
## Why
Managed marketplace source requirements only become effective when every
local marketplace mutation path applies the same admission decision.
This change centralizes that decision so CLI, app-server, and
external-agent migration flows cannot add, install from, or refresh a
disallowed source.
## What changed
- Match exact normalized Git repository URLs with an optional exact
`ref`.
- Match Git hosts with managed regular expressions.
- Match local marketplaces by exact absolute path.
- Preserve the expected path/name boundary for managed OpenAI
marketplaces.
- Enforce source admission during marketplace add, plugin install, and
configured Git marketplace upgrade.
- Continue upgrading independent marketplaces when one source is
rejected and return a per-marketplace error.
- Load the effective requirements stack at CLI, app-server, and
external-agent migration entry points.
This PR does not filter already configured marketplaces at runtime; that
remains in draft follow-up #29691.
## Stack
This is PR 2 of 3 and is based on #29690, which introduces the
requirements data shape and merge behavior.
## Test plan
- Source matcher coverage for Git URL/ref, host-pattern, local-path, and
managed marketplace cases.
- Marketplace add and plugin install coverage for allowed and rejected
sources.
- Marketplace upgrade coverage for rejection and per-marketplace
continuation.
## Why
Authentication mode is a domain concept used by login, model selection,
telemetry, and transports. Keeping the canonical type in app-server
protocol forces those lower-level crates to depend on an unrelated wire
API.
## What changed
- Added canonical `codex_protocol::auth::AuthMode` domain values.
- Kept the app-server wire DTO unchanged and added an explicit app-side
conversion.
- Removed production app-server-protocol dependencies from login,
model-provider-info, models-manager, and otel call paths.
## Stack
This is PR 2 of 6, stacked on [PR
#29714](https://github.com/openai/codex/pull/29714). Review only the
delta from `codex/split-json-rpc-protocols`. Next: [PR
#29722](https://github.com/openai/codex/pull/29722).
## Validation
- Auth and login coverage passed in the focused protocol/domain test
run.
- App-server account and auth conversion coverage passed.
## Why
Fork-specific response items, including the subagent usage hint, are
appended directly to `InitialHistory::Forked`. This bypasses the normal
history insertion path that assigns missing response item IDs when
`Feature::ItemIds` is enabled, so the child could reconstruct and
persist those items without IDs.
## What changed
- When `Feature::ItemIds` is enabled, assign missing IDs to top-level
`ResponseItem`s while materializing `InitialHistory::Forked`, before
both reconstruction and persistence.
- Preserve existing IDs and use the same owned rollout items for live
history and persistence.
- Extract the existing single-item ID allocation logic for reuse by the
fork path.
- Add coverage that verifies a fork-only developer message receives the
same ID in live and persisted history with the feature enabled.
Normal history recording, compacted-history replacement, and fork
handling all continue to honor `Feature::ItemIds`. External-agent
imports, normal resume, and nested legacy compaction checkpoints are
unchanged.
## Testing
- `just test -p codex-core
record_initial_history_reconstructs_forked_transcript`
- `just test -p codex-core
record_initial_history_assigns_and_persists_id_for_forked_response_item`
## Why
Managed deployments need a mergeable way to declare which marketplace
sources Codex may use. An enterprise-keyed TOML table avoids array merge
ambiguity and lets every requirements layer use the existing config
precedence rules without a marketplace-specific merger.
## Requirements shape
```toml
[marketplaces]
restrict_to_allowed_sources = true
[marketplaces.allowed_sources.company_plugins]
source = "git"
url = "https://github.com/example/company-plugins.git"
ref = "main"
[marketplaces.allowed_sources.internal_git]
source = "host_pattern"
host_pattern = "^git\\.example\\.com$"
[marketplaces.allowed_sources.local_plugins]
source = "local"
path = "/opt/company/codex-plugins"
```
`restrict_to_allowed_sources` follows normal scalar precedence.
`allowed_sources` follows normal recursive TOML table merge behavior:
distinct keys accumulate and fields under the same key use normal layer
precedence. The final `source` value later selects which fields the
marketplace admission policy interprets.
The raw rule fields remain optional while requirements layers are
composed, so a higher-priority layer can override only `ref`, `url`, or
another individual field. Source-specific validation and normalization
intentionally belong to the marketplace admission layer, not
requirements merging.
This initial shape includes `git`, `host_pattern`, and `local` sources.
It does not add npm or path-pattern rules.
## What changed
- Add the marketplace requirements TOML shape to
`ConfigRequirementsToml`, `ConfigRequirementsWithSources`, and
`ConfigRequirements`.
- Carry marketplace requirements through the existing regular
requirements merge path.
- Keep allowed-source entries as raw partial tables for downstream
policy interpretation.
- Cover partial same-key overlays, source changes, unknown fields, and
unmodified local paths.
This PR defines and composes the requirements only. Source admission is
implemented by the next PR in the stack.
## Stack
This is PR 1 of 3. #29753 adds source admission on top of this PR; draft
#29691 will add runtime enforcement after it is rebased later.
## Test plan
- `just test -p codex-config marketplace_`
## Why
`start_new_context_window` independently replaced in-memory history and
persisted a compacted checkpoint instead of using the shared
compacted-history path. That bypassed the centralized missing-item-ID
assignment when `item_ids` is enabled, so fresh context messages could
enter the new context window and its persisted replacement history
without IDs.
This follows up on the token-budget compaction reset flow introduced in
[#29743](https://github.com/openai/codex/pull/29743).
## What changed
- Delegate new context-window installation to
`replace_compacted_history`.
- Reuse its ID assignment, in-memory replacement, world-state baseline,
checkpoint persistence, turn-context persistence, and session-start
bookkeeping.
- Add focused coverage that verifies generated IDs are present in live
history and preserved in the persisted replacement history.
## Testing
- `just test -p codex-core
start_new_context_window_assigns_and_persists_item_ids`
- `just test -p codex-core
new_context_tool_starts_new_window_before_follow_up`
## Why
Some extension hosts need generated images returned without writing them
to the local filesystem or giving the model a local path.
## What changed
**tl;dr**: we now conduct all extension operations in the image gen
extension
- Let hosts provide an optional image save root when installing the
extension.
- Save images and return path hints only when a save root is configured.
- Return image data without saving or adding a path hint when no save
root is configured.
- Preserve the extension-provided `saved_path` instead of persisting
extension images again in core.
- Leave built-in image generation unchanged.
## Validation
- `just test -p codex-image-generation-extension`
- `just test -p codex-app-server
standalone_image_generation_returns_saved_path_hint_to_model`
- `just test -p codex-core
extension_tool_uses_granted_turn_permissions_without_local_persistence`
- `just test -p codex-core tools::handlers::extension_tools::tests`
- tested on CODEX CLI on both save_root: CODEX_HOME and None
- tested on CODEX APP on both as well
## Why
Start moving towards app-server tests defaulting to running against
remote & foreign OS executors. To do so we need a point of indirection
similar to core integration tests' `build_with_auto_env`, but with the
flexibility of letting tests control environment registration if they
need to.
## What
This adds:
- `TestAppServer::new_with_auto_env()` for constructing an app server
with a default environment defined by the test runner (e.g. bazel)
- `TestAppServer::auto_env_params()` for tests to easily acquire turn
env params tailored to the automatic environment
- `TestAppServer::send_thread_start_request_with_auto_env()` to make it
easy for tests to start a thread using the automatic environment
The above methods all fail if the test calling them has set up an
environment where the automatic environment configuration conflicts with
test-created state.
## Validation
Adds a couple of basic smoke tests to the app-server test suite.
Follow-ups will migrate more tests to use it.
## Why
The `ItemIds` path fills in missing IDs before response items are
persisted and emitted as raw item events. `ResponseItem::AgentMessage`
is part of that same response-item stream, but it was skipped by the
missing-ID repair path, leaving agent messages without stable item IDs
while messages and tool items received generated IDs.
Agent messages recorded through `InterAgentCommunication` also need the
generated ID to survive rollout persistence and resume. Otherwise
clients can observe an `amsg_` ID for the live raw response item, then
see that same persisted agent message lose its item ID after restart.
## What changed
- Assign missing `ResponseItem::AgentMessage` IDs with the `amsg_`
prefix.
- Persist the generated item ID on `InterAgentCommunication` and replay
it back into the reconstructed `ResponseItem::AgentMessage` on resume.
- Keep the persisted ID out of the model-visible inter-agent message
envelope.
- Keep `CompactionTrigger` and `Other` skipped because they do not get
generated item IDs.
- Update session/protocol tests for agent-message ID assignment and
resume preservation.
## Manual Testing
Run the local dev build using `just c --enable item_ids` to ensure this
code is exercised:
https://github.com/openai/codex/blob/322e33512b2d38d38d705e2ef692a8aca50decac/codex-rs/core/src/session/mod.rs#L2713-L2715
In the `.jsonl` file, I saw entries like:
```json
{
"timestamp": "2026-06-24T00:44:03.098Z",
"type": "inter_agent_communication",
"payload": {
"id": "amsg_019ef715-849a-7a50-becc-ce63c6a9c994",
```
## Test plan
- `just test -p codex-core
record_inter_agent_communication_preserves_item_id_in_rollout_and_resume`
- `just test -p codex-core
record_inter_agent_communication_sets_turn_id_in_rollout_and_resume`
- `just test -p codex-protocol
inter_agent_communication_response_input_item_preserves_commentary_phase`
## Why
With `DeferredExecutor`, a sampling request can begin while an
environment is still starting. The model can see that pending state, but
needs a way to wait for the environment within the same turn before
continuing.
Environment startup is owned by Core, so the wait tool should use the
same request-frozen `StepContext` that advertised the starting
environment. This keeps tool registration and execution tied to the
exact startup operation the model saw, even if live thread state later
changes.
Supersedes #29735.
## What
- register `wait_for_environment` when the current `StepContext`
contains starting environments
- wait on the selected `StartingTurnEnvironment` shared resolution and
return a bounded ready or failed result
- rebuild the next request normally, removing the wait tool and exposing
ready environment tools, or reporting the environment as unavailable
after failure
## Testing
- `just test -p codex-core deferred_executor_`
- verifies the wait tool is replaced by environment-backed tools after
startup
- verifies startup failure removes both the wait tool and unavailable
environment tools while notifying the model
## Why
Work(TPP) threads can be launched from the Desktop app, but if they all
keep the Desktop app's default originator then downstream attribution
cannot distinguish local Work launches from cloud-backed Work launches.
`thread/start.serviceName` already carries that launch signal, while
`SessionMeta.originator` is the durable thread-level value that survives
resume and fork.
This change converts the Desktop Work service names into an effective
originator at thread creation time, persists that originator with the
thread, and keeps using it for later model requests and memory writes.
## What changed
- Map `CODEX_WORK_LOCAL` and `CODEX_WORK_CLOUD` service names to
per-thread originators, while preserving
`CODEX_INTERNAL_ORIGINATOR_OVERRIDE` as the highest-precedence override.
- Persist the effective originator in `SessionMeta.originator`, read it
back on resume/fork, and inherit the parent originator for subagent
spawns when there is no persisted session metadata.
- Handle truncated `SpawnAgentForkMode::LastNTurns` forks by falling
back to the live parent originator when the forked history no longer
includes `SessionMeta`.
- Thread the per-thread originator through Responses headers,
websocket/compaction request paths, thread-store creation, rollout
metadata, and memory stage-one telemetry.
## Verification
- `just test -p codex-core
agent::control::tests::spawn_thread_subagent_inherits_parent_originator_without_fork
agent::control::tests::spawn_thread_subagent_fork_last_n_turns_inherits_parent_originator_without_session_meta
thread_manager::tests::originator_override_precedes_service_name_remapping`
- `just test -p codex-core
agent::control::tests::resume_thread_subagent_restores_stored_metadata_and_effective_multi_agent_mode`
- `just test -p codex-memories-write`
- `just fix -p codex-core -p codex-memories-write`
- `git diff --check`
## Why
When `Feature::TokenBudget` is enabled, compaction should behave like
`new_context`: start a fresh context window with the standard injected
context, without asking the server to summarize old history and without
carrying prior user or assistant messages into the next model request.
This is still a compaction operation from the client lifecycle
perspective. Manual `/compact` and auto-compaction should keep the same
observable side effects that clients and hooks expect, including compact
hooks and `TurnItem::ContextCompaction`.
## What changed
- Added `compact_token_budget` to run token-budget manual and inline
auto-compaction through a shared compaction lifecycle.
- Split pending `new_context` requests from forced context-window
startup: `take_new_context_window_request()` consumes pending requests,
and `start_new_context_window()` installs a fresh context window.
- Routed token-budget manual `/compact` and inline auto-compaction to
install a fresh context window locally instead of calling server/local
summarization.
- Preserved compact lifecycle side effects for token-budget compaction
by running pre/post compact hooks and emitting `ContextCompaction` item
start/completion events.
- Updated token-budget tests to assert fresh window IDs, absence of
server-side compaction calls, dropped prior transcript messages/tool
output after reset, and compact hook/item lifecycle behavior.
## Testing
- `just test -p codex-core
token_budget_context_uses_new_window_after_compaction`
- `just test -p codex-core token_budget_compaction_runs_compact_hooks`
- `just test -p codex-core
token_budget_mid_turn_auto_compaction_resets_before_active_follow_up`
---------
Co-authored-by: pakrym-oai <pakrym@openai.com>
## Summary
- rename the rollout-budget exhaustion error from
`RolloutBudgetExceeded` to `SessionBudgetExceeded`
- expose the matching app-server v2 wire value as
`sessionBudgetExceeded`
- regenerate JSON/TypeScript schema fixtures and update the app-server
docs and focused tests
This is a naming-only follow-up to #29715 based on [Pavel's review
suggestion](https://github.com/openai/codex/pull/29715#discussion_r3463183480).
Runtime behavior is unchanged.
## Tests
- `just test -p codex-core rollout_budget`
- `just test -p codex-app-server-protocol`
- `just fmt`
- `just write-app-server-schema`
## Why
With `model_auto_compact_token_limit_scope = "body_after_prefix"`, the
persistent prefix should not count against the active body window.
`get_context_remaining` and the token-budget reminder should report the
same usable body-after-prefix window that auto-compaction uses, rather
than the total token count since the session began.
This is stacked on #29664 so the mechanical move from `turn.rs` is
isolated from the behavior fix.
## What
- Extends `ContextWindowTokenStatus` with `context_remaining_tokens`.
- Updates `get_context_remaining` to use the shared context-window
accounting.
- Adds integration coverage for body-after-prefix reminder timing and
`get_context_remaining` output.
## Testing
- `just test -p codex-core body_after_prefix_window`
- `just test -p codex-core auto_compact_body_after_prefix`
- `just fix -p codex-core`
Owen LinandGitHub
rka-oaiandGitHub
Michael BolinandGitHub
Celia ChenandGitHub
jifandGitHub
sayan-oaiandGitHub
AdrianandGitHub
Ahmed IbrahimandGitHub
Shijie RaoandGitHub
Alex DaleyandGitHub
Adam Perry @ OpenAIandGitHub
DrewandGitHub
TomandGitHub
Brent TrautandGitHub
rhan-oaiandGitHub
xl-openaiandGitHub
pakrym-oaiandGitHub
Won ParkandGitHub
alexsong-oaiandGitHub
Andrey MishchenkoandGitHub