Skip to main content

Health Check

System status and monitoring endpoints. No authentication required.

GET /api/health​

Basic system health check.

Response:

{
  "status": "operational",
  "uptime": "2h 30m 45s",
  "uptimeSeconds": 9045,
  "memory": {
    "used": "45MB",
    "total": "128MB",
    "rss": "156MB"
  },
  "providers": {
    "available": ["openai", "anthropic", "google", "grok", "openrouter"],
    "count": 5
  },
  "models": {
    "total": 38
  },
  "cache": {
    "size": 150,
    "valid": 120,
    "expired": 30
  },
  "timestamp": "2026-01-17T18:00:00.000Z"
}

GET /api/health/status​

Detailed status of each provider with connectivity test.

Cache: 30 seconds

Response:

{
  "status": "healthy",
  "providers": {
    "openai": {
      "provider": "openai",
      "status": "healthy",
      "latency": 254,
      "tested": true,
      "lastCheck": "2026-01-17T18:00:00.000Z",
      "model": "gpt-5-mini"
    },
    "anthropic": {
      "provider": "anthropic",
      "status": "healthy",
      "latency": 198,
      "model": "claude-sonnet-4-5-20250929"
    }
  },
  "summary": {
    "total": 5,
    "healthy": 5,
    "degraded": 0,
    "averageLatency": 320
  }
}

GET /api/health/metrics​

Detailed system metrics including CPU, memory, and performance.

Cache: 60 seconds

Response:

{
  "status": "operational",
  "system": {
    "platform": "linux",
    "nodeVersion": "v20.10.0",
    "pid": 1234,
    "cpu": {
      "user": 1234567,
      "system": 234567
    },
    "loadAverage": [0.5, 0.4, 0.3]
  },
  "uptime": "2h 30m 45s"
}

POST /api/health/check​

Force a complete health check without using cache.

Response: Same as GET /api/health/status with "forced": true