Loop Nanny

Monitor, triage, and clean up after agent loops. The agent-loop skill starts loops. This skill keeps them healthy.

Monitor a Running Loop

joelclaw loop status <LOOP_ID>

joelclaw is the primary CLI (igs is a legacy alias for the same binary).

Poll every 2–3 minutes. A story typically takes 3–8 minutes (test-write → implement → review → judge). If a story shows no progress for 10+ minutes, something is stuck.

Reading Status Output

Skip Triage

Stories skip when the judge fails them after max retries. But the implementation often landed anyway — later stories may have included the work, or the first attempt was correct but tests were over-specified.

Check if skipped work actually landed

# 1. Read the attempt output
cat /tmp/agent-loop/<LOOP_ID>/<STORY_ID>-<ATTEMPT>.out | tail -40

# 2. Check if the files exist on main after merge
ls <project>/path/to/expected/file

# 3. Check git log for the story's commits
cd <project> && git log --oneline --grep="<STORY_ID>" -5

Common skip causes

Post-Loop Cleanup

After merge-back completes (status shows RETRO.COMPLETED then resolves):

1. Verify merge landed

cd <project> && git log --oneline -5
# Should show: "Merge branch 'agent-loop/<LOOP_ID>'"

2. Run full test suite

cd <project> && bun test 2>&1 | tail -5

3. Delete stale acceptance tests

Agent loops generate __tests__/<story-id>-*.test.ts files. These test implementation details and break on every refactor. Delete them after verifying the real tests pass.

# Find agent-generated test files
ls <project>/__tests__/*-*.test.ts 2>/dev/null

# Check which fail
bun test __tests__/ 2>&1 | grep "(fail)"

# Delete ALL agent-generated acceptance tests (they served their purpose)
rm <project>/__tests__/<prefix>-*.test.ts

# Verify clean
bun test 2>&1 | tail -3

Per AGENTS.md: “over-specified tests that mock internal step names are worse than no tests — they break on every refactor and give false negatives.”

4. TypeScript check

cd <project> && bunx tsc --noEmit

5. Restart worker (if loop touched Inngest functions)

launchctl kickstart -k gui/$(id -u)/com.joel.system-bus-worker
sleep 3
joelclaw refresh
joelclaw status

6. Commit cleanup

cd <project> && git add -A && git commit -m "chore: post-loop cleanup for <LOOP_ID>

Delete stale acceptance tests, verify N pass / 0 fail / tsc clean"

When to Intervene

Don’t intervene — let the loop work

• Story on first attempt (even if slow)
• Story retrying with feedback (judge gave actionable feedback)
• Story just skipped but later stories are still pending

Intervene — something is broken

• Same error on retry: Check attempt output. If retry 2 has identical error to retry 1, the feedback loop isn’t helping — cancel and fix the root cause.
• Merge conflict during complete: joelclaw logs errors will show merge abort. Manually resolve: cd <project> && git merge --abort then merge the branch by hand.
• Worker crashed: joelclaw status shows worker down. Restart: launchctl kickstart -k gui/$(id -u)/com.joel.system-bus-worker
• All stories skipping: The PRD likely has a bad assumption. Cancel, review prd.json, re-fire.
• Loop stuck (no progress for 15min): Check joelclaw runs --count 5 — if the latest run is COMPLETED but no new run dispatched, there’s a state bug. Cancel and re-fire from the stuck story.

Cancel a stuck loop

joelclaw loop cancel <LOOP_ID>

# Clean up worktree manually (cancel doesn't auto-merge)
cd <project>
git worktree remove /tmp/agent-loop/<LOOP_ID> --force 2>/dev/null
git branch -D agent-loop/<LOOP_ID> 2>/dev/null
git worktree prune

Reading Attempt Output

Each story attempt writes to /tmp/agent-loop/<LOOP_ID>/<STORY_ID>-<ATTEMPT>.out. These contain the implementor/reviewer’s full output (diffs, reasoning, test results).

# Quick scan — last 40 lines usually has the verdict
tail -40 /tmp/agent-loop/<LOOP_ID>/<STORY_ID>-<ATTEMPT>.out

# Check all attempts for a story
ls /tmp/agent-loop/<LOOP_ID>/<STORY_ID>-*.out

# Grep for errors across all attempts
grep -i "error\|fail\|reject" /tmp/agent-loop/<LOOP_ID>/*.out

Monitoring Checklist

Use this sequence when babysitting a loop:

1. joelclaw loop status <LOOP_ID>          — where are we?
2. (if story running) wait 3 min, re-check
3. (if story skipped) check attempt output — did the work land anyway?
4. (if loop completed) verify merge, run tests, delete stale tests, restart worker
5. (if stuck) joelclaw runs --count 5 + joelclaw logs errors — diagnose

Improve joelclaw While You Nanny

The nanny is the primary consumer of joelclaw output. When you hit a gap — missing info, unclear output, an extra step you had to do manually — fix joelclaw right then. The joelclaw package is at ~/Code/joelhooks/joelclaw/ (safe to edit while a loop runs against the worker clone).

What to improve

• Missing data in output: joelclaw loop status doesn’t show story descriptions? Add them.
• Manual steps that should be commands: Had to kubectl logs directly? Add it to joelclaw logs.
• Bad next_actions: The suggested next commands don’t match what you actually needed? Fix the HATEOAS.
• Error without fix: Got an error with no fix field? Add one via respondError().
• New command needed: Found yourself running raw curl/GraphQL? Wrap it in a joelclaw command.

How to improve

# Commands are in src/commands/ — one file per command
# Response helpers in src/response.ts
# Inngest client methods in src/inngest.ts

# Test your change
joelclaw <command> 2>&1 | python3 -m json.tool

# Commit
git add -A && git commit -m "feat: joelclaw <command> — <what you added>"

Follow the cli-design skill: JSON always, HATEOAS next_actions, context-safe output, errors with fixes. After improving joelclaw, update this skill doc if the monitoring workflow changed.

Diagnosing Stalled Loops

When a loop stops progressing, use the diagnostic command FIRST:

# Quick diagnosis of all loops
igs loop diagnose all -c

# Diagnose + auto-fix (clears claims, re-fires plan events)
igs loop diagnose all -c --fix

# Verify fix worked (~30s later)
igs loop status <loop-id> -c

Common diagnoses:

• CHAIN_BROKEN — judge→plan event was lost. --fix re-fires the plan event.
• ORPHANED_CLAIM — agent died but claim remains. --fix clears claim + re-fires.
• STUCK_RUN — Inngest run active but agent dead. --fix clears + re-fires.
• WORKER_UNHEALTHY — fewer functions than expected. --fix restarts worker.

See loop-diagnosis skill for full reference.

Gotchas

• Don’t edit the monorepo while a loop runs — git add -A in the worktree scoops unrelated changes
• joelclaw IS safe to edit while a loop runs — the worker clone is separate from the monorepo
• prd.json and progress.txt get dirty — this is normal; complete.ts stashes before merge
• Worktree branch not auto-deleted on cancel — clean manually (see Cancel section)
• Cross-file test pollution — agent-generated __tests__/ files can cause failures in real tests when run together; delete them