swarm-self-heal

Swarm reliability watchdog for OpenClaw — validates gateway/channel and every lane, performs bounded recovery, and emits auditable receipts.

This skill appears to implement the described watchdog behavior and uses only the OpenClaw CLI plus bash/jq, but review a few things before installing: 1) Inspect setup.sh and change/remove the hard-coded Telegram fallback '8563003761' (or ensure your $HOME/.openclaw/openclaw.json contains the correct channel target) to avoid sending alerts to an unexpected recipient. 2) Installing will copy scripts into ~/.openclaw/workspace-studio/scripts and register cron jobs that run autonomously and may restart the user-level openclaw-gateway service—make a backup of OpenClaw cron/jobs.json and openclaw.json first. 3) Run the scripts manually in a controlled environment (or review outputs from check.sh) to confirm behavior before enabling cron. If you want higher assurance, ask the author for an explanation of the fallback number or a configurable opt-in for external notifications.

Type: OpenClaw Skill Name: swarm-self-heal Version: 0.1.1 The 'swarm-self-heal' skill is designed to monitor and self-heal an OpenClaw agent swarm. It uses standard OpenClaw CLI commands, `bash`, and `jq` to perform health checks, restart the gateway if needed, and ping agents. Persistence is established via cron jobs in `scripts/setup.sh`, which is a core, stated feature for a watchdog. The skill utilizes prompt engineering in `scripts/setup.sh` to instruct the agent running the cron job to execute a local script and format its output, and in `scripts/swarm_self_heal.sh` to send benign 'ping' messages to other agents. While these are instances of prompt injection, they are used for the skill's stated purpose and show no evidence of malicious intent, data exfiltration, or unauthorized actions. All operations are aligned with system maintenance and reliability.