Monitoring

The MCPBox dashboard provides at-a-glance monitoring of your servers, tools, request volume, and error rates.

The Activity page provides detailed execution history and protocol logs for debugging.

Execution history with success/failure status, duration, and tool name.

MCP protocol logs showing request types, levels, and messages.

Health Checks

MCPBox provides several health check endpoints:

# Basic health
curl http://localhost:8000/health
# → {"status": "healthy"}

# Detailed health with service status
curl http://localhost:8000/health/detail

# Individual service health
curl http://localhost:8000/health/services

The detailed health check reports the status of PostgreSQL, the sandbox, and circuit breakers.

Prometheus Metrics

MCPBox exposes a /metrics endpoint (enabled by default).

curl http://localhost:8000/metrics

Key metrics:

Metric	Type	Description
http_request_duration_seconds	Histogram	Request latency (p50, p95, p99)
http_requests_total	Counter	Request count by method, handler, status
http_request_size_bytes	Histogram	Request body sizes
http_response_size_bytes	Histogram	Response body sizes

To disable metrics, set ENABLE_METRICS=false in .env.

Webhook Alerting

Configure ALERT_WEBHOOK_URL in .env to receive critical alerts:

# Discord
ALERT_WEBHOOK_URL=https://discord.com/api/webhooks/your-webhook-id/your-token

# Slack
ALERT_WEBHOOK_URL=https://hooks.slack.com/services/your/webhook/url

Alerts fire on circuit breaker trips, security events, and other critical conditions.

Logs

MCPBox logs to stdout in structured JSON format.

# View all logs
docker compose logs -f

# View specific service
docker compose logs -f backend

# View last 100 lines
docker compose logs --tail=100 backend

Configure your Docker logging driver for retention:

# docker-compose.override.yml
services:
  backend:
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "3"