watchdog

Detect stale leases and requeue stalled work items in a single transactional commit.

Updated Mar 9, 2026
One-click install
npx skills add https://github.com/ether-moon/agent-atelier --skill watchdog-ether-moon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: watchdog
Source: https://github.com/ether-moon/agent-atelier/tree/main/plugins/agent-atelier/skills/watchdog
Command: npx skills add https://github.com/ether-moon/agent-atelier --skill watchdog-ether-moon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Health check and mechanical recovery — detect stale leases, expired candidates, stuck reviews, and blocked work items, then take safe recovery actions. It also enforces operating budgets and replays interrupted transactions. Use for routine health checks, when something seems stuck, or when the user says 'watchdog', 'health check', 'check for stale items', 'anything stuck?', 'run maintenance', 'check budgets', 'recovery sweep', or 'clean up orphaned leases'. Also appropriate after a session crash or long idle period.

Core Features & Use Cases

  • Mechanical recovery actions: requeue and reset lease state for stale work items, clear ownership fields, and bump requeue counters.
  • Budget and safety enforcement: verify effort against thresholds, emit budget_exceeded alerts, and replay interrupted transactions to preserve consistency.
  • Observability & escalation: generate alerts for long-open gates, repeated failures, and orchestrator review triggers; support safe respawn and re-dispatch.

Quick Start

Run the watchdog tick to perform a health check and recovery sweep.

Frequently Asked Questions about watchdog

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate recovery for stalled agent-workflows with expired leases?

Automate recovery for stalled agent-workflows by running a health check tick that targets lease expiry, stale reviews, and candidate timeouts. It safely requeues work items, clears ownership fields, and bumps requeue counters in a single transactional commit.

What is the best way to enforce operating budgets and replay interrupted transactions in state-management?

Enforce operating budgets in state-management by verifying effort against defined thresholds during a watchdog tick. The process emits budget_exceeded alerts and replays interrupted transactions to preserve consistency within a single commit.

How do health checks handle orphaned leases and stuck reviews during a recovery sweep?

Health checks handle orphaned leases and stuck reviews by mechanically resetting lease state, clearing ownership fields, and generating escalation alerts for long-open gates or repeated failures during the recovery sweep.

Can I trigger a maintenance recovery sweep after a session crash or long idle period?

Trigger a maintenance recovery sweep after a session crash or long idle period by running the watchdog tick. It reads and mutates loop state via state-commit to detect stale items and safely re-dispatch blocked work.

When should I not use an automated health check for mechanical recovery?

Avoid automated health check recovery when manual intervention is required for complex state conflicts or when work items are still actively processing, as the tick targets only mechanically stalled leases and expired candidates.