Lingxi · Agent Health Monitor

lingxi-agent-monitor

Proactive health monitoring and self-healing for AI Agent systems built on OpenClaw.

What It Monitors

Usage

Trigger Commands

"System health check"  → Full health scan + report
"Agent status"        → Quick status summary
"Memory check"         → Focus on memory metrics
"Task queue check"     → Focus on pending/failed tasks
"Alert me if..."      → Configure custom alert rules

Health Report Format

# Agent Health Report — YYYY-MM-DD HH:MM

## System Status
| Component | Status | Latency | Notes |
|-----------|--------|---------|-------|
| OpenClaw Gateway | ✅ Healthy | 45ms | |
| MCP Server | ✅ Connected | 120ms | |
| GitHub Integration | ✅ OK | 200ms | |
| ZeroGPU | ✅ OK | 80ms | |

## Session Overview
- Active sessions: 3
- Avg session duration: 23m
- Memory: 1.2GB / 4GB (30%)

## Recent Activity
- Tasks completed: 47
- Tasks failed: 2 (4.3%)
- Avg task duration: 1m 12s

## Alerts (if any)
- [ ] No critical alerts
- ⚠️ Warning: ZeroGPU latency > 500ms (3 occurrences)

Alert Levels

Self-Healing Actions

When issues are detected, the agent can automatically:

1. Restart failed MCP connections
2. Clean up stale sessions (sessions idle > 2 hours)
3. Flush memory cache when usage > 80%
4. Retry failed tool calls (up to 3 attempts)
5. Switch to backup providers (e.g., GitHub API fallback)

Configuration

{
  "alert_channels": ["log", "notification"],
  "memory_threshold_pct": 80,
  "session_idle_timeout_min": 120,
  "max_sessions": 20,
  "error_rate_threshold_pct": 10,
  "latency_threshold_ms": 5000,
  "self_heal_enabled": true,
  "health_check_interval_min": 15
}

Cron Schedule

*/15 * * * * → Health check every 15 minutes
0 */2 * * *  → Deep memory cleanup every 2 hours

Installation

clawhub install lingxi-agent-monitor

Integration with External Alerting

For external alerts (Slack, Discord, PagerDuty):

# In your notification workflow
curl -X POST $SLACK_WEBHOOK \
  -H "Content-Type: application/json" \
  -d '{"text": "🔴 Agent Alert: $ALERT_MESSAGE"}'

Memory Management

When memory usage is high:

1. Archive completed session summaries to memory files
2. Clear intermediate computation cache
3. Merge small memory files into consolidated ones
4. Prune session history beyond retention limit

Dashboard Endpoint

If running with n8n bridge:

GET /agent-monitor/dashboard  → HTML dashboard with real-time status
GET /agent-monitor/metrics   → JSON metrics endpoint
GET /agent-monitor/alerts     → JSON alert history