Repository navigation
Keep Antigravity background-task handoffs active while Core replays them - #1385
Conversation
A turn that ends while agy's own background task still runs was parked as interrupted/usage_limit_wait, so the mission looked stopped for the ~80 s before its automatic replay (mission 2e3bf791). Give the handoff its own terminal reason (background_task_handoff) and keep the mission Active; the replayer, recovery payload and Stop handling recognise both reasons.
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 24fec26e8a
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| if reason == BACKGROUND_HANDOFF_REASON { | ||
| MissionStatus::Active |
There was a problem hiding this comment.
After an Antigravity background-task handoff, keeping the mission Active prevents its automatic continuation: replay_remote_mission_after_usage_limit calls continue_on_node, whose continue_inner rejects placement.live || mission.status == MissionStatus::Active with REMOTE_JOB_STILL_RUNNING (src/api/control/remote_grok.rs:3180–3197), even when the previous node job is terminal and its lease has been finished. The replayer therefore retries every five minutes and eventually marks the mission failed instead of continuing it. Update that dispatch guard to distinguish this replayable handoff from an actually running job, and exercise the real continuation path in the regression test.
Useful? React with 👍 / 👎.
A turn that ends while agy's own background task still runs was parked as
interrupted/usage_limit_wait, so the mission looked stopped for the ~80 s
before its automatic replay (mission 2e3bf791). Give the handoff its own
terminal reason (background_task_handoff) and keep the mission Active; the
replayer, recovery payload and Stop handling recognise both reasons.