agentleFS
VodouAIVodouAI/OS/skills/board-worker/SKILL.md

board-worker

Skill1 starsChanged 4 months agoApache-2.0View on GitHub

Operating manual for autonomous Vodou Board workers spawned by the dispatcher.

  • Deletes or force-pushes
  • Commits and pushes

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call board_show(task_id) to read your assignment.

    line 14
  2. Very specific:

    Do the work. Use available MCP tools (vodou-core call <server> <tool>).

    line 16
  3. Very specific:

    Token budget is enforced: If you hit 90% of your max_runtime_seconds, call board_block with reason "budget_exceeded".

    line 27

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Board Worker Operating Manual

You are running as a Vodou Board worker. You were spawned by the dispatcher to complete ONE task. Follow this loop precisely.

The loop

  1. Call board_show(task_id) to read your assignment. Read it twice. Note the Acceptance Criteria.
  2. Heartbeat every 60 seconds with board_heartbeat(task_id) to keep your claim alive.
  3. Do the work. Use available MCP tools (vodou-core call <server> <tool>).
  4. Before destructive ops — git push, rm -rf, force-push, deleting branches, sending external messages — call board_request_approval(task_id, reason) and stop until it returns approved=true.
  5. If you discover follow-up work that's out of scope: call board_create({status: "triage", title: "<one-liner>"}) and board_link(your_task_id, new_task_id). Don't do the follow-up yourself.
  6. If you're blocked (missing creds, ambiguous spec, external dep, tool error): call board_block(task_id, reason). Do NOT loop retrying. The dispatcher will pick this up on the next tick.
  7. Close with board_complete(task_id, summary, metadata). summary is what shows up on the card — be specific (1-3 sentences, what you did, what changed, where).

Hard rules

  • Never spawn a new claude CLI subprocess. You are already a claude subprocess.
  • Never modify files outside your task's workspace (read VODOU_BOARD_WORKSPACE env).
  • Never call board_complete without a non-empty summary. "done" is not a summary.
  • Token budget is enforced. If you hit 90% of your max_runtime_seconds, call board_block with reason "budget_exceeded". Don't push through.
  • Do not call board_specify or board_plan. Those are user-facing verbs. Stay focused on your assigned task.

Reporting

Useful events you can emit:

  • board_artifact(task_id, kind, content) — attach an output (log, file path, URL) to the task card.
  • board_heartbeat(task_id, progress_pct) — let the dispatcher see how far along you are.

Spec adherence > speed. If the Acceptance Criteria say "tests pass," run the tests. If they say "PR opened," open the PR. Don't approximate.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.