execute-bead-worker

name	execute-bead-worker
description	Execute one assigned bead inside a swarm coordinated by swarm-epic. Use only when a coordinator has assigned the bead, file scope, and verification contract; the worker implements, verifies, and reports without mutating bead state.

Execute Bead Worker

Implement one assigned bead inside a swarm run.

Goal

Deliver one bead safely inside the boundaries set by swarm-epic.

Steps

1. Confirm the assignment from the coordinator:
   • epic id
   • bead id
   • Files: scope
   • Verify: commands or checks
   • reservation or conflict instructions
2. Read the bead details and inspect the relevant code.
3. If the bead is underspecified or missing Files or Verify, stop and return it to the coordinator instead of guessing.
4. Register the worker and inspect its inbox:
   • Windows:

     .\scripts\windows\agent-mail.ps1 --repo . register --name worker/<bead-id> --role worker --epic-id <epic-id> --bead-id <bead-id>
     .\scripts\windows\agent-mail.ps1 --repo . inbox --recipient worker/<bead-id>
     

   • POSIX:

     ./scripts/posix/agent-mail.sh --repo . register --name worker/<bead-id> --role worker --epic-id <epic-id> --bead-id <bead-id>
     ./scripts/posix/agent-mail.sh --repo . inbox --recipient worker/<bead-id>
     

5. Reserve the declared file scope before editing through the shared control plane:
   • Windows:

     .\scripts\windows\agent-mail.ps1 --repo . reserve --owner worker/<bead-id> --epic-id <epic-id> --bead-id <bead-id> --path <path1> --path <path2>
     

   • POSIX:

     ./scripts/posix/agent-mail.sh --repo . reserve --owner worker/<bead-id> --epic-id <epic-id> --bead-id <bead-id> --path <path1> --path <path2>
     

   If reservation fails, stop and report the conflict to the coordinator.
6. Update .beads/workflow/HANDOFF.json for this worker context:
   • role
   • epic_id
   • bead_id
   • summary
   • next_action
7. Post a started message to bead/<bead-id> so the coordinator and other sessions can see who owns the bead.
   • Windows:

     .\scripts\windows\agent-mail.ps1 --repo . post --thread bead/<bead-id> --sender worker/<bead-id> --type started --body '{"status":"started"}' --epic-id <epic-id> --bead-id <bead-id>
     

   • POSIX:

     ./scripts/posix/agent-mail.sh --repo . post --thread bead/<bead-id> --sender worker/<bead-id> --type started --body '{"status":"started"}' --epic-id <epic-id> --bead-id <bead-id>
     

8. Implement only within the assigned scope.
9. Run the assigned verification commands and any required repo-local build-and-test checks for the touched surface area.
10. Report back to the coordinator with:

• changed files
• verification commands run
• key output or exit status
• any new risks or follow-up work
• confirmation that reservations were released

11. Release reservations and post either a completed or blocked message to bead/<bead-id>.

• release syntax:
  • Windows: .\scripts\windows\agent-mail.ps1 --repo . release-reservations --owner worker/<bead-id> --bead-id <bead-id>
  • POSIX: ./scripts/posix/agent-mail.sh --repo . release-reservations --owner worker/<bead-id> --bead-id <bead-id>

12. If blocked or context-limited:

• update HANDOFF.json
• release reservations if possible
• report the blocker clearly to the coordinator

Hard Rules

• Do not run bd update, bd close, or any other bead status mutation.
• Do not expand the file scope without coordinator approval.
• Do not silently skip verification.
• Do not keep reservations after you stop working.
• Do not assume another session can see local .beads/workflow/; shared coordination only happens through Agent Mail.