claude-lead-andresmgsl/heavy-duty-watcher

grok-bot-andresmgsl 1d00544e86 Initial release: org-wide heavy-duty GitHub review watcher

Polls all heavy-duty/* repos every 15m for assignee/review-requested
work, reviews with clear Verdicts until agreement, and includes
tmux restore + hourly health self-heal. Fully documented.

2026-07-18 17:40:05 +00:00

1.7 KiB

Raw Blame History

Operations runbook

First-time setup

# 1. Clone
git clone https://github.com/grok-bot-andresmgsl/heavy-duty-watcher.git
cd heavy-duty-watcher

# 2. Auth
gh auth status   # must be grok-bot-andresmgsl (or your reviewer bot)

# 3. Install live + start tmux loops
./scripts/install-live.sh

# 4. In a Grok session: arm durable schedulers (15m poll + 1h health)
#    Or restore interactive session:
./scripts/restore.sh

After PC reboot / box restart

gh auth status
~/heavy-duty-watcher/scripts/restore.sh
# or:
~/heavy-duty-watcher/scripts/restore.sh --status

Day-to-day

# What is waiting for us right now?
./scripts/discover.sh

# One-shot health
./scripts/health-check.sh

# Follow activity
tail -f logs/poll.log logs/health.log

# Attach tmux
tmux attach -t heavy-duty-watcher

Schedulers (Grok)

When the agent is live, it should keep two durable recurring tasks:

Poll every 15 minutes — org-wide discovery + review
Health every 1 hour — health-check.sh + self-heal poll-loop

If either disappears, re-create from the prompts in docs/SCHEDULER_PROMPTS.md.

Migration from `~/rig-watcher`

The original single-repo install lived at ~/rig-watcher.
install-live.sh can symlink ~/rig-watcher → ~/heavy-duty-watcher for compatibility.
Item keys changed from pr:N to heavy-duty/rig#pr:N.

Troubleshooting

Symptom	Check
No reviews happening	`./scripts/discover.sh` empty? Not requested.
Stale last_poll	`health-check.sh` WARN; is poll-loop alive?
gh 401	`gh auth login`
Duplicate spam	Inspect `state.json` `last_head_sha` / `last_action`
Scheduler missing	Re-arm via Grok; see `docs/SCHEDULER_PROMPTS.md`

1.7 KiB Raw Blame History