## Why
Shared hosts serve independent chats and their subagents. Larger callback workloads need room for unrelated work and control messages, while gRPC completion bursts can exhaust HTTP/2's small-DATA-frame budget and disconnect shared sessions.
## What changed
- Raise the active-cell limit from 128 to 2,048 and the pending delegate callback limit from 1,024 to 4,096. Expand request and transport queues to leave room for completion bursts and control traffic.
- Negotiate larger callback budgets through the stdio `delegate-capacity-4096` capability and gRPC `max_pending_delegate_calls` field. Preserve the 1,024-callback budget for legacy clients and enforce both session and shared host limits on gRPC.
- Limit concurrent gRPC completion RPCs to 64 across a provider's sessions, allowing queued completions to be canceled without waiting for delivery capacity.
## Testing
Add stdio and gRPC integration tests covering 256 cells and 4,096 callbacks across shared sessions while unrelated work completes. Add tests for callback budget isolation and reuse, rejection of zero capacity, and cancellation of queued completions. Include an ignored resource-intensive test for full host capacity.
GitOrigin-RevId: e85bfa24cc46cc8a735b464d80adf8066a6a6d3a
## Why
Code Mode sessions always used client-side output handling, leaving server truncation settings and observation budgets unused.
## What changed
- Honor each session's `output_truncation` setting in the runtime and both host transports, and advertise the bounded-output capability.
- Apply `max_output_tokens` independently to execute, wait, and terminate responses in server mode. Keep client mode as the default when the setting is omitted.
- Use bounded output collection and rendering in server mode, preserving head and tail output with omission notices. Include script errors in the output budget while retaining failure status and a separate diagnostic capped at 64 KiB.
- Estimate audio tokens without the Tokio-dependent cache when rendering bounded output.
## Testing
Add runtime and host regression tests for session mode selection, per-request budgets, output overflow, oversized images, zero-token audio output, diagnostic preservation, and termination under output backpressure.
GitOrigin-RevId: 3d9661edcdd2dde31526bb92eb83fa0614f0fa88
## What changed
- Add `GetProcessMetrics` and provider sampling APIs for process ID, start time, cumulative CPU time, and resident memory, exposed through `ThreadManager` without creating a thread or session.
- Distinguish unsupported collection, a locally absent host, and sampling failures. Process-owned providers never start or replace a host when sampling.
- Bound gRPC sampling to one second without retries or explicit connection resets. Run optional collectors on a detached worker with at most one collection in flight, so stalled collection cannot block the HTTP/2 driver or runtime shutdown.
## Testing
Add coverage for snapshots, collector errors and panics, deadlines, overload, and shutdown with a stuck collector. Extend lifecycle tests to check that sampling preserves session state and keepalive behavior without starting or restarting local hosts.
GitOrigin-RevId: eeafe275891d0464636f28a29345a87aed1a12dc
## What changed
Add session options for client- or server-owned output truncation and optional `max_output_tokens` budgets for wait and terminate observations. Introduce `TerminateRequest` and update session implementations and callers to use it.
Extend the stdio and gRPC schemas with optional truncation settings and budgets, and define the `bounded-output-v1` capability. Existing sessions retain client-side truncation; providers reject server mode by default. This change defines the contracts without enabling server-side truncation.
## Testing
Extend protocol serialization coverage for `outputTruncation: "server"` while preserving existing wire shapes when optional fields are absent.
GitOrigin-RevId: 22145897ea16d385c5ba9e7549109f7a669fceb4
## What changed
- Add `grpc+stdio://` as a host transport and the disabled-by-default `code_mode_host_grpc` feature flag to select it.
- Share a lazy HTTP/2 channel and host process across code-mode sessions while keeping session state isolated.
- Use HTTP/2 keepalives to detect unresponsive hosts, kill and reap disconnected children, and spawn a fresh host on reconnect. Exit the host when its stdio connection closes.
## Testing
Add coverage for shared-host session isolation, host exit on stdin EOF, and recovery after an unresponsive host is reaped. Exercise nested tool calls with both legacy stdio and gRPC stdio, asserting that the feature flag selects the expected transport.
GitOrigin-RevId: 31389d8540de236b8e0571d3c24e29e117b75c4a
## What changed
Add an optional `preempt` signal to `CodeModeSession::execute` and `CodeModeSession::wait` so callers can end an observation early while the cell continues running and remains available for later waits.
Support the signal in the in-process runtime and across both remote transports: `operation/yield` with the negotiated `yield-observation` capability for stdio, and `YieldObservation` for gRPC. Preserve signals received before gRPC observation registration, and retain normal timeout behavior for older hosts that lack support.
## Testing
Add runtime and transport tests covering early execute and wait yields followed by successful cell completion, yields before observation registration, and suppression of yield frames for older stdio hosts.
GitOrigin-RevId: 3f8049d8f0582be6b8f0832cc03f2d974a0b8f18
## Why
A delegate bound at session creation cannot provide different callbacks for cells sharing that session. Each execution needs to retain its own delegate across yields and release it when the cell is cleaned up.
## What changed
- Pass `CodeModeSessionDelegate` to `execute` instead of session creation.
- Route tool calls, notifications, and cell closure callbacks through the execution's delegate in the in-process runtime and the gRPC and stdio transports.
- Retain delegates with pending executions and live cells, releasing them through closure and cancellation cleanup.
## Testing
Add coverage for distinct delegates across yielded cells, gRPC callbacks before cell admission, and delegate release after completion or abandoned execution cleanup. Update transport tests to verify callbacks reach the owning cell's delegate.
GitOrigin-RevId: 7469f52104b993e790a40c65fb800ad02b7fd606
## What changed
- Add `codex-otel-trace-websocket` with a `TraceWebSocket` API that owns the
loopback OTLP receiver and WebSocket listener.
- Bind both listeners during startup, expose the exporter and bound listener
addresses, and surface listener failures through a single lifecycle method.
- Update `codex-code-mode-host` to use the new crate and shut down the bridge
after flushing its trace provider.
GitOrigin-RevId: ec5ca4c4369b6b3b1232875c2699730eb748ab1c
## Why
Code mode wall time should measure the host operation itself, without including
client-side response delays or idle time between requests.
## What changed
- Measure each execute, wait, and terminate request in the code mode host.
- Carry the duration through the stdio and gRPC protocols and use it for
model-visible wall time.
- Emit a structured `codex.code_mode.host_timing` event correlated with the
conversation, turn, tool call, and cell.
## Testing
- Cover successful and failed execution timing, delayed response reads,
repeated waits, termination, and missing cells across stdio and gRPC.
- Verify timing survives protocol serialization and is reflected in app-server
model output and structured telemetry.
GitOrigin-RevId: d24af30c3fc5820521b4beba1f9970714dad6482
## What changed
- Add `--otel-trace-exporter` for sending OTLP/HTTP JSON traces directly to an exporter.
- Add the mutually exclusive `--otel-trace-listen` option for streaming raw OTLP trace batches over a trace-only WebSocket, backed by a loopback OTLP receiver.
- Instrument host startup, transport requests, and gRPC operations with request, session, execution, cell, wait, and tool-call identifiers where available.
- Flush the trace provider with a bounded timeout during shutdown.
GitOrigin-RevId: 4dcdb2bc3b15166dccd5041fb875caea74f3f6fe
## What changed
- Accept only `http://` and `https://` gRPC endpoints for app server
`--code-mode-host` connections.
- Limit the standalone code-mode host to stdio and gRPC listeners.
- Remove the WebSocket session provider, dual-WebSocket negotiation, transport
lanes, and related dependencies and tests.
- Keep coverage for shared and prewarmed gRPC hosts and reject WebSocket host
URLs in CLI parsing tests.
GitOrigin-RevId: 8792604e79ffd54ce776074a4dc3f7d55e63752a
## What changed
- Accept `grpc://IP:PORT` endpoints through `--listen` and serve the existing
code-mode gRPC service over TCP.
- Print the bound HTTP endpoint to stdout so callers can discover the port when
binding to port `0`.
- Apply the protocol frame-size limits and disable Nagle's algorithm on accepted
connections.
## Testing
- Add an end-to-end test that starts the host on an ephemeral port, connects a
gRPC client, and opens a session.
- Verify accepted gRPC sockets have `TCP_NODELAY` enabled.
GitOrigin-RevId: 51d6c21dff8cffb47068c0677ba10ff370385cec
## What changed
- Export `GrpcCodeModeHost` as a transport-independent implementation of the
code-mode gRPC API.
- Support leased sessions, execution and wait lifecycle operations, filtered
nested tool-call subscriptions, tool completions, and notification
acknowledgements.
- Share host-wide request and active-cell limits across the existing and gRPC
transports, and bound identifiers, metadata, subscriptions, and pending
callbacks.
## Testing
- Add coverage for request conversion, ordered callback routing, cancellation,
session cleanup, backpressure, malformed input, and resource-limit handling.
GitOrigin-RevId: f146ba7e6fe4e4aa02f25dd3f961120980516d0e
## What changed
- Add `create_session_with_limits` and session-scoped cell execution limits.
- Clamp execute and wait yield times to the session's `max_yield_time_ms`
without terminating the running cell.
- Negotiate support with remote code-mode hosts and include non-default limits
in `session/open`, while keeping unlimited sessions compatible with hosts and
providers that do not support limits.
## Testing
- Cover yield-time clamping, zero-timeout behavior, and isolation between
sessions.
- Cover wire serialization, capability negotiation, unsupported hosts, and
shared process-host execution.
GitOrigin-RevId: 9517321cd605bb87f93eeaa6ba331cc2e346e582
## Why
Large nested-tool callbacks can occupy a WebSocket and delay unrelated session
operations on the same code-mode connection.
## What changed
- Negotiate the optional `dual-websocket-v1` capability and pair a second,
token-scoped WebSocket with the control connection.
- Route nested-tool callbacks and their results over the bulk socket while
keeping session operations, notifications, and execution responses on the
control socket. Reject messages sent on the wrong lane.
- Preserve the single-connection transport when the capability is unavailable,
and bound pairing, queued callbacks, and deferred cross-socket messages.
- Defer callbacks that arrive before their execution-started response, and
return delegate errors without disconnecting the connection.
## Testing
Add protocol, transport, driver, and WebSocket integration coverage for
capability negotiation, lane routing, pairing failures, out-of-order messages,
and progress during large concurrent tool results.
GitOrigin-RevId: fa4504653e7cbf3c4ec930ae57aa0a41345bad66
## What changed
- Move the V8 implementation into a dedicated `codex-code-mode-runtime` crate used by `codex-code-mode-host`, removing the embedded runtime fallback from the Codex process.
- Resolve the host executable from the active installation layout and check its availability before selecting tools.
- Fall back to direct tools with a one-time warning when optional code mode is unavailable. Keep `code_mode_only` and `disable_in_process_fallback` configurations fail-closed.
## Testing
- Cover host discovery for standalone and package layouts, including missing hosts and symlinks.
- Verify direct-tool fallback, one-time warnings, and fail-closed code-mode-only behavior.
GitOrigin-RevId: 5aa3c6f1db148b2231fc24089a2ee0e2b00dbddb
## What changed
- Add a `--listen` option that accepts `stdio`, `stdio://`, or a
`ws://IP:PORT` endpoint, while retaining stdio as the default.
- Serve the existing length-prefixed protocol in binary WebSocket messages,
with isolated connections, shared host limits, and a `/readyz` endpoint.
- Reject browser-origin handshakes and contain malformed frames to the affected
connection.
## Testing
- Cover listen URL parsing and complete-frame encoding and decoding.
- Exercise readiness, cell execution, tool callbacks, large frames, concurrent
connections, malformed frames, and origin rejection through the WebSocket
listener.
GitOrigin-RevId: 01c8be4c6256b8ce4a3a0002440dcb3294e5f887
## Summary
- implement the standalone `codex-code-mode-host` stdio service
- route sessions, cells, delegate requests, responses, and cancellation
through a bounded host peer
- supervise request, writer, cell-forwarding, actor, and V8 failure
boundaries
- bound request/session tombstones and fail-stop the connection on
invalid protocol state
- add host-only duplex protocol tests and local Cargo/Bazel run recipes
## Why
This stage makes the host process independently runnable and reviewable
before exposing any remote client in Codex. Transport or runtime failure
closes the connection and relies on process replacement rather than
transactional recovery.
## Stack
This is **3 of 4** in the process-owned code-mode session stack.
- Depends on #30110
- The final client PR targets this branch
## Validation
- `just test -p codex-code-mode-host` — 7 host-only tests passed
- `just fix -p codex-code-mode-host`
- `just bazel-lock-update`
- `just bazel-lock-check`
- `just fmt`
pash-openai