Monitor agent health with dead-man switch
Automated monitoring of agent or cron job status with alerts and state capsules if the agent becomes unresponsive.
How it works
Changes data in your accountsOperations and infrastructure
The agent registers a monitor for a specific cron job or autonomous agent using the Flight Recorder server. It sets up heartbeats to ensure the agent is alive. If the agent fails to send a heartbeat, the system triggers an alert with the last state capsule and a resume link. The agent must ensure the monitor is correctly registered and that the key allows for the required number of monitors.
Prompt
Paste this into an agent that has the servers below connected. Adjust the names and dates to your own.
Prompt for Monitor agent health with dead-man switch
Register a monitor for my daily backup cron job and set up a heartbeat. If it fails, alert me with the last state capsule.MCP servers it needs
Source
The vendor page this playbook is drawn from, and the passage that supports it.
A dead-man switch for autonomous agents and cron jobs: heartbeats with a state capsule; silence triggers an alert with the last capsule and a resume link.AIS Gateway — the MCP campus for your AI assistant