System Health Auditor

Overview

This agent executes a critical, mandatory 'Heartbeat Protocol' upon every wake cycle. Its primary function is to perform a deep health audit of the entire operational environment, ensuring all connected services and internal configurations are functioning correctly before any major tasks begin.

It follows a strict sequence: reviewing past lessons learned, checking the immediate inbox for alerts, and running multiple structured API calls to validate system components.

Capabilities

• Inbox Monitoring: Checks the agent's direct inbox for any pending issues or critical messages.
• CLI Version Validation: Verifies the installed versions of core development CLIs (Claude, Codex, Gemini) to ensure compatibility.
• Agent Configuration Audit: Scans all registered agents to validate that adapter types and configured models are compatible with each other (e.g., checking if a 'claude' adapter is incorrectly pointing to a 'gemini' model).
• Budget Monitoring: Identifies any agent whose monthly budget usage has exceeded an 80% threshold, flagging potential overspending.
• Status Check: Audits for agents stuck in error states or those that have been running without recent updates.

Example Use Cases

• Pre-Deployment Checklist: Run this immediately before launching a complex workflow to guarantee all dependencies are healthy.
• Scheduled Nightly Routine: Integrate this into a cron job to provide an automated, non-disruptive system health report every night.
• Troubleshooting Baseline: When experiencing intermittent failures, running the Heartbeat Protocol provides a clean baseline report of what should be working versus what is actually failing.