PR #22579 real-runtime verification
main: 15fe283272; PR head: 282206f98e4520a734c9f76eb8e82b19f71d755c
Linux, system Chrome 151, repository Playwright, real Codex gpt-6.1-sol via local proxy.
App-created project /home/user/queue-verification; real git repository and turns.
Delegated tasks requested through authenticated orchestration.dispatchCommand with completionWake=settled_only, to exercise the queue rather than active steering. ProviderContinuationService generated the completion notices. No state writes, fixtures, mocks or injected provider responses.
Selection changed through thread.model-selection.set over the same WebSocket RPC used by the app/MCP.

MAIN FAILURE
2026-10-31T20:09:18.974Z seq=418 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:09:44.060Z seq=424 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:10:49.238Z seq=529 run.created run=10 queued {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:10:58.235Z seq=536 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:11:50.495Z seq=557 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:11:50.495Z seq=560 run.updated run=10 starting {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:11:58.681Z seq=579 run.updated run=10 completed {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}

PR SUCCESS + USER QUEUE BOUNDARY
2026-10-31T20:16:39.352Z seq=130 run.created run=3 queued {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:16:49.359Z seq=136 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:17:43.817Z seq=157 run.updated run=3 starting {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:17:46.987Z seq=175 run.updated run=3 completed {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:20:54.351Z seq=180 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:21:23.809Z seq=288 run.created run=6 queued {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:21:24.091Z seq=293 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:22:16.136Z seq=316 run.updated run=6 starting {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:22:24.544Z seq=336 run.updated run=6 completed {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:22:27.179Z seq=343 run.created run=7 starting {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:22:32.030Z seq=366 run.updated run=7 completed {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:22:34.282Z seq=383 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:24:47.859Z seq=424 run.created run=10 queued {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:24:48.114Z seq=429 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:25:02.391Z seq=432 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"low"}]}
2026-10-31T20:25:35.088Z seq=453 thread.model-selection-updated  {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:25:35.088Z seq=456 run.updated run=10 starting {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}
2026-10-31T20:25:38.329Z seq=474 run.updated run=10 completed {"instanceId":"codex","model":"gpt-6.1-sol","options":[{"id":"reasoningEffort","value":"high"},{"id":"serviceTier","value":"default"}]}

Real provider completion: API_CHILD_READY. Next user reply: NEXT_USER_LOW / PARENT_SPAWN_WINDOW (context preserved). Queued user reply: USER_SELECTION_HIGH.
Blacksmith: 177/177 passed across runtimeLayer, DelegatedCompletionDelivery, SteeringCompletion.integration, ProviderContinuationService, QueuedRunOrder, testkit/ProviderSwitch.integration, ClaudeAutomaticDelivery.integration, EffectWorker.
No source changes.