This server pipes Netdata's real-time infrastructure metrics directly into Claude. It connects to Netdata Cloud's API and exposes per-second monitoring data, anomaly detection alerts, and ML-powered insights from your infrastructure. You'd reach for this when troubleshooting production issues or investigating performance patterns and want Claude to analyze live metrics, correlate alerts across services, or explain anomalies detected by Netdata's edge-based models. It supports streamable HTTP transport, so you get continuous metric updates rather than one-off snapshots. Useful if you're already running Netdata agents on your infrastructure and want to query their collected data through natural language instead of navigating dashboards or writing queries.
claude mcp add --transport stdio io.github.netdata-mcp-server uvx mcp-server