health

Health Check Report

Perform a health check on all agents and identify any issues.

Instructions

1. Get all agents using mcp__trinity__list_agents

2. For each running agent, check:

   • Container status (should be "running")
   • Context usage (warn if >90%, critical if >95%)
   • Last activity timestamp (warn if >30 minutes ago for running agents)

3. Classify each agent's health:

   • Healthy: Running, context <75%, recent activity
   • Warning: Context 75-90%, or idle >30 minutes
   • Critical: Context >90%, container errors, or stuck

4. Generate a health report with sections:

## Agent Health Report
Generated: {timestamp}

### Overall Health
{Overall status: Healthy / Degraded / Critical}

### Critical Issues (Immediate Attention Required)
{List any critical issues, or "None"}

### Warnings
{List any warnings, or "None"}

### Healthy Agents
{Count of healthy agents}

### Detailed Status

| Agent | Health | Issue | Recommendation |
|-------|--------|-------|----------------|
| ... | ... | ... | ... |

Health Criteria

Condition	Level	Action
Context >95%	Critical	Suggest session reset
Context 90-95%	Warning	Monitor closely
Context 75-90%	Warning	Consider reset soon
Container not running	Critical	Restart if expected to run
No activity >30 min (running)	Warning	Check if stuck
No activity >60 min (running)	Critical	Likely stuck, restart

Save Report

After generating the report:

1. Create directory if needed: mkdir -p ~/reports/health
2. Save with timestamp: ~/reports/health/YYYY-MM-DD_HHMM.md
3. Confirm: "Report saved to ~/reports/health/YYYY-MM-DD_HHMM.md"

Notes

• Skip trinity-system (yourself) from health checks
• Focus on actionable issues, not just status
• Provide specific recommendations for each issue