You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In town 6b110d52-9a70-4742-91d8-fab68a2bec3c (rig nx.ts), the pr-feedback and rework bead dispatchers do not check the current state of the target PR before creating or dispatching new beads.
After the merge, pr-feedback beads with title Address review comments on PR #18 (nx.ts) continued to be created and dispatched. Polecats correctly identified the PR as merged and closed the bead as no-op, but a new bead was immediately created.
Over ~2 hours post-merge, at least 4 such beads accumulated, all dispatch_attempts: 5, all failed as no-op.
Expected: before creating a new pr-feedback or rework bead, the dispatcher (or patrol service) should query the target PR via gh api graphql ... pullRequest.state. If state is MERGED or CLOSED, the dispatcher should NOT create a new bead. The webhook-driven thread creation should also check state.
Related: the pr-feedback bead template includes a pre-flight gh pr view step. The bead is created before any pre-flight can run. So the gating needs to be in the dispatcher, not the bead.
Severity: medium. Wastes polecat compute, pollutes dashboard, distracts mayor from real work.
Area
Agent Dispatch / Scheduling
Context
Town ID: 6b110d52-9a70-4742-91d8-fab68a2bec3c
Agent: Mayor (e920f2fe-6fce-4534-ae2e-43549fd8c66f)
Rig ID: 905bf756-546a-484c-9b4e-1a0b6b903457
Recent Errors
Recurring stale pr-feedback dispatch on merged PR #18 — 4th no-op. PR merged at 10:44Z, beads continued dispatching through 16:00Z.
Filed automatically by the Mayor via gt_report_bug.
What happened?
In town
6b110d52-9a70-4742-91d8-fab68a2bec3c(rignx.ts), the pr-feedback and rework bead dispatchers do not check the current state of the target PR before creating or dispatching new beads.Reproduction:
https://github.com/ThePlenkov/nx.ts/pull/18) was MERGED at 2026-06-23T10:44:29Z.Address review comments on PR #18 (nx.ts)continued to be created and dispatched. Polecats correctly identified the PR as merged and closed the bead as no-op, but a new bead was immediately created.dispatch_attempts: 5, all failed as no-op.Expected: before creating a new pr-feedback or rework bead, the dispatcher (or patrol service) should query the target PR via
gh api graphql ... pullRequest.state. If state isMERGEDorCLOSED, the dispatcher should NOT create a new bead. The webhook-driven thread creation should also check state.Related: the pr-feedback bead template includes a pre-flight
gh pr viewstep. The bead is created before any pre-flight can run. So the gating needs to be in the dispatcher, not the bead.Severity: medium. Wastes polecat compute, pollutes dashboard, distracts mayor from real work.
Area
Agent Dispatch / Scheduling
Context
Recent Errors
Filed automatically by the Mayor via
gt_report_bug.