Grafana MCP
About
Grafana's official MCP server lets agents reach your observability stack in natural language — query Prometheus/Loki metrics and logs, search dashboards, view and silence alerts, investigate incidents, bringing monitoring data into AI troubleshooting.
Our Verdict
Grafana MCP brings monitoring capabilities to the MCP ecosystem. Its GitHub popularity reflects community trust, and the standardized protocol ensures broad compatibility across AI clients.
Pricing
Use Cases
- Integrating monitoring capabilities into AI agent workflows
- Building custom AI-powered tools with Grafana MCP
- Enhancing AI agent functionality with standardized MCP access
Pros
- Grafana MCP provides standardized MCP protocol access
- Open source with 3,305 GitHub stars
- Active development and community support
Cons
- Requires MCP-compatible client (Claude, Cursor, etc.)
- Setup complexity varies by client
- Some features may depend on client capabilities
Installation
uvx mcp-grafanaFrequently Asked Questions
What can Grafana MCP help me do?
It brings Grafana observability into your AI assistant: query Prometheus metrics, search Loki logs, find dashboards and panels, view/silence alerts and investigate incidents in natural language. Instead of jumping across pages while troubleshooting, just ask "which host had the recent CPU spike".
How do I install Grafana MCP and what token is needed?
The recommended path is a zero-install uvx mcp-grafana run (a Docker image and prebuilt binaries are also provided). Set GRAFANA_URL and a Service Account Token (created in Grafana) via environment variables, granting read-only scopes for metrics, logs and alerts as needed.
Is Grafana MCP free?
The MCP server is officially open-sourced by Grafana Labs and free (Apache-2.0); it accesses your own Grafana instance. Self-hosted Grafana OSS costs nothing, and Grafana Cloud has a free tier with usage billed per its plans beyond that.