agentleFS
Sign inStart free
zoranmarticzoranmartic/coord-wright/skills/coord-loop-doctor/SKILL.md

coord-loop-doctor

Skill0 starsChanged 47 days agoApache-2.0View on GitHub

Diagnose stuck, quiet, token-heavy, or repeatedly requeued coord loops. Use when coord-loop seems unreliable, tasks sit in needs-brainstorming or *-working too long, wrapper state looks stale, or the user asks why coord is not moving.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill before starting another expensive loop.

    line 8
  2. Distinctive:

    When diagnosing coord behavior, read these first:

    line 12
  3. Very specific:

    Read task-files-reference.md only when authoring or transitioning tasks.

    line 18
  4. Very specific:

    Check the project's state files and log: .coord/worker.state, .coord/worker.log.

    line 24
  5. Distinctive:

    Check launchd service state, recent logs, and provider retry markers.

    line 25
  6. Very specific:

    Check whether the task is genuinely runnable or blocked by depends_on.

    line 26
  7. Distinctive:

    Check whether main checkout dirt or fast-forward failure prevented pickup.

    line 27
  8. Very specific:

    Use coord-status, coord-requeue, coord-unblocker, and coord-tokens only when they match the diagnosis.

    line 28

≈ “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

Coord Loop Doctor

Use this skill before starting another expensive loop. The goal is one diagnosis and the smallest safe recovery action.

Required Reading

When diagnosing coord behavior, read these first:

  1. ${COORD_TOOLS:-$HOME/Projects/coord-wright}/docs/coordination.md
  2. ${COORD_TOOLS:-$HOME/Projects/coord-wright}/docs/codex-coordination.md or claude-coordination.md, depending on the stuck agent
  3. ${COORD_TOOLS:-$HOME/Projects/coord-wright}/docs/task-files.md

Read task-files-reference.md only when authoring or transitioning tasks.

Workflow

  1. Resolve the canonical project root. Do not mutate queue state from sibling task worktrees.
  2. Inspect the handoff or task with the smallest useful coord view: show --handoff, then show --compact, then scoped files only if needed.
  3. Check the project's state files and log: .coord/worker.state, .coord/worker.log.
  4. Check launchd service state, recent logs, and provider retry markers.
  5. Check whether the task is genuinely runnable or blocked by depends_on.
  6. Check whether main checkout dirt or fast-forward failure prevented pickup.
  7. Use coord-status, coord-requeue, coord-unblocker, and coord-tokens only when they match the diagnosis.

Classification

Return exactly one primary classification:

  • healthy
  • queue-starved
  • dependency-blocked
  • provider-limited
  • stale-wrapper
  • stuck-task
  • dirty-main
  • needs-human

Use Grill-Me

Invoke grill-me when recovery would mutate task state, when multiple worktrees could own recovery, or when the fix is a wrapper change rather than a one-task requeue.

Output

Report classification, evidence, smallest recovery action, and whether a follow-up task should be shaped.

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.

Check your setup, free