Grafana MCP Server

Instructions

Implementation Reference

• src/grafana_mcp_server/processor/grafana_processor.py:592-605 (handler)

  The core implementation that makes the HTTP request to the Grafana API to fetch datasources.

  def grafana_fetch_datasources(self) -> dict[str, Any]:
      """
      Fetches all datasources from Grafana.
  
      Returns:
          Dict containing list of datasources
      """
      try:
          url = f"{self.__host}/api/datasources"
          logger.info("Fetching all datasources")
  
          response = requests.get(url, headers=self.headers, verify=self.__ssl_verify, timeout=20)
  
          if response.status_code == 200:

• src/grafana_mcp_server/mcp_server.py:432-446 (handler)

  The wrapper function in the MCP server that interfaces with the processor to call the tool.

  def grafana_fetch_datasources():
      """Fetch all datasources from Grafana"""
      try:
          grafana_processor = current_app.config.get("grafana_processor")
          if not grafana_processor:
              return {
                  "status": "error",
                  "message": "Grafana processor not initialized. Check configuration.",
              }
  
          result = grafana_processor.grafana_fetch_datasources()
          return result
      except Exception as e:
          logger.error(f"Error fetching datasources: {e!s}")
          return {"status": "error", "message": f"Failed to fetch datasources: {e!s}"}

• src/grafana_mcp_server/mcp_server.py:263-267 (registration)

  The tool definition/registration in the MCP server list.

  {
      "name": "grafana_fetch_datasources",
      "description": "Fetches all datasources from Grafana with their configuration details.",
      "inputSchema": {"type": "object", "properties": {}, "required": []},
  },