How do I install the Watchclaw skill?

Run npx skillmds@latest add dvcrn/watchclaw in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Watchclaw safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: executes scripts, makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Watchclaw?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Watchclaw free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Watchclaw?

dvcrn (@dvcrn) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Watchclaw by dvcrn — Agent Skill

Name: Watchclaw
Rating: 4.2 (5 reviews)
Author: dvcrn

File contents

watchclaw

OpenClaw gateway watchdog — auto-recovery from bad config changes.

What It Does

watchclaw monitors your OpenClaw gateway and automatically recovers from bad configurations:

Health polling — checks gateway HTTP endpoint every N seconds
Config change detection — detects uncommitted or new commits in your config repo
Auto-recovery — stashes uncommitted changes (U1) or reverts bad commits (U2) via git
Probation — validates stability after config changes before promoting to known-good
Pluggable alerts — iMessage, webhook, or custom command on failure/recovery

Usage

# Start watching (background daemon)
watchclaw --config /path/to/watchclaw.conf start

# Start in foreground (for debugging)
watchclaw --config /path/to/watchclaw.conf start --foreground

# Check status
watchclaw --config /path/to/watchclaw.conf status

# Follow logs
watchclaw --config /path/to/watchclaw.conf logs -f

# Stop
watchclaw --config /path/to/watchclaw.conf stop

Config

Create a .conf file (see watchclaw.conf.example):

GATEWAY_PORT=18790
GATEWAY_CONFIG_DIR="$HOME/.openclaw"
POLL_INTERVAL_SEC=10
HEALTH_TIMEOUT_SEC=5
GATEWAY_TLS=0              # Use https for health check
MAX_RETRIES=3
ALERT_HOOK="imsg"           # imsg | webhook | command | none
ALERT_IMSG_TO="user@me.com"
RESTART_MODE="native"       # native | docker

Recovery Modes

Scenario	Detection	Recovery
Uncommitted config change breaks gateway	Health check fails + dirty openclaw.json	`git stash` → restart
Bad commit breaks gateway	Health check fails + new commit	`git revert` → restart
Config change during healthy operation	HEAD ≠ known-good in HEALTHY state	Enter probation, monitor

Docker Mode

For containerized OpenClaw (e.g., 飞书/Feishu bot):

RESTART_MODE="docker"
DOCKER_CONTAINER="openclaw-feishu"

Requirements

bash 4+, git, curl
OpenClaw gateway config must be in a git repo
python3 or node for JSON validation

---name: watchclawdescription: Auto-recovery watchdog for OpenClaw gateway. Monitors health, detects bad config changes, and recovers via git stash/revert. Supports native and Docker restart modes with pluggable alerts.---# watchclawOpenClaw gateway watchdog — auto-recovery from bad config changes.## What It Doeswatchclaw monitors your OpenClaw gateway and automatically recovers from bad configurations:- **Health polling** — checks gateway HTTP endpoint every N seconds- **Config change detection** — detects uncommitted or new commits in your config repo- **Auto-recovery** — stashes uncommitted changes (U1) or reverts bad commits (U2) via git- **Probation** — validates stability after config changes before promoting to known-good- **Pluggable alerts** — iMessage, webhook, or custom command on failure/recovery## Usage```bash# Start watching (background daemon)watchclaw --config /path/to/watchclaw.conf start# Start in foreground (for debugging)watchclaw --config /path/to/watchclaw.conf start --foreground# Check statuswatchclaw --config /path/to/watchclaw.conf status# Follow logswatchclaw --config /path/to/watchclaw.conf logs -f# Stopwatchclaw --config /path/to/watchclaw.conf stop```## ConfigCreate a `.conf` file (see `watchclaw.conf.example`):```bashGATEWAY_PORT=18790GATEWAY_CONFIG_DIR="$HOME/.openclaw"POLL_INTERVAL_SEC=10HEALTH_TIMEOUT_SEC=5GATEWAY_TLS=0 # Use https for health checkMAX_RETRIES=3ALERT_HOOK="imsg" # imsg | webhook | command | noneALERT_IMSG_TO="user@me.com"RESTART_MODE="native" # native | docker```## Recovery Modes| Scenario | Detection | Recovery ||----------|-----------|----------|| Uncommitted config change breaks gateway | Health check fails + dirty openclaw.json | `git stash` → restart || Bad commit breaks gateway | Health check fails + new commit | `git revert` → restart || Config change during healthy operation | HEAD ≠ known-good in HEALTHY state | Enter probation, monitor |## Docker ModeFor containerized OpenClaw (e.g., 飞书/Feishu bot):```bashRESTART_MODE="docker"DOCKER_CONTAINER="openclaw-feishu"```## Requirements- `bash` 4+, `git`, `curl`- OpenClaw gateway config must be in a git repo- `python3` or `node` for JSON validation

Frequently asked questions

Watchclaw