About
MCP server for Grafana dashboards, datasources, queries, and observability workflows.
Features
- Connect MCP clients to Grafana instances with service-account tokens
- Summarize and patch dashboards without sending full dashboard JSON
- Query Prometheus, Loki, InfluxDB, ClickHouse, CloudWatch, Graphite, Athena, Snowflake, and Elasticsearch/OpenSearch through Grafana datasources
- Inspect dashboards, datasources, and query examples from an AI client
- Run via uvx, Docker, binary, or Helm according to the official README
Use Cases
SRE
Ask an MCP client to inspect dashboards, run panel queries, and explain incidents without leaving the chat workflow.
Platform engineer
Expose Grafana observability context to coding agents while keeping authentication in Grafana service-account tokens.
What This Server Can Do
MCP servers expose three types of capabilities to AI clients. Here's what Grafana MCP Server supports.
Tools
Supported18 toolsFunctions your AI client can call to perform actions — like querying a database, creating a file, or calling an API.
How to use: Tools run automatically when your AI client decides they're needed. Ask your AI assistant to perform a task, and it will invoke the right tool.
Resources
Not availableThis server does not expose read-only data resources.
Prompts
Not availableThis server does not provide pre-built prompt templates.
For the full list of available tools, resources, and prompts, check the README on GitHub.
Available Tools (3)
dashboard_summaryGet compact dashboard summaries with title, panels, variables, and metadata.
patch_dashboardApply targeted changes to Grafana dashboards.
query_datasourceRun datasource queries through Grafana-supported datasources.