Pilot Heartbeat Monitor

Detect agent failures and trigger automatic task redistribution or re-election. Use this skill when: 1. You need to detect when swarm members become unreacha...

This skill appears to do what it says (heartbeat publication and failure detection) but there are gaps you should clarify before installing or running it: - Ensure pilotctl is a trusted binary on PATH and you understand how it authenticates to the registry (the SKILL.md assumes you can publish and read messages but doesn't declare required credentials or how to provide them). - The instructions use environment variables ($AGENT_ID, $SWARM_NAME, $REGISTRY_HOST) that are not declared in the skill metadata—decide how you will set these and whether any of them contain sensitive values. - Confirm the network and privacy implications of publishing heartbeats to the registry and of pinging peers (these actions expose agent IDs and addresses on the network). - Verify the presence and provenance of tooling the scripts call (jq, bc) and run the commands in a restricted/test environment first. - Note the skill is licensed AGPL-3.0; if you embed or modify it, that license may affect redistribution of your modified skill. If you need a cleaner security posture, request the author to: declare required env vars and any credentials, document authentication flows for pilotctl/registry, and state expected network/trust boundaries.

Type: OpenClaw Skill Name: pilot-heartbeat-monitor Version: 1.0.0 The pilot-heartbeat-monitor skill provides standard bash-based logic for agent health monitoring and failure detection using the 'pilotctl' utility. The code in SKILL.md focuses on publishing heartbeats, parsing message history with jq, and verifying peer connectivity via pings, all of which align with its stated purpose without any signs of malicious intent, data exfiltration, or unauthorized execution.