About Grafana MCP SSE Server
Grafana MCP SSE Server connects MCP-compatible clients to Grafana for dashboard and panel queries. It supports dashboard search, panel listing, and Prometheus panel data queries.
Key features
- 01Search dashboards by name
- 02List dashboard panels
- 03Query Prometheus panel data
Use cases
- 01Find dashboards related to a service
- 02Inspect panels in a dashboard
- 03Retrieve Prometheus panel metrics
MCP support
An MCP server using SSE that connects MCP clients to the Grafana API. It exposes dashboard search, panel listing, and panel data queries as MCP capabilities.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README📊 aimiox Grafana MCP
The aimiox MCP Grafana SSE Server provides an easy way to query Grafana dashboards and panels using the Model Context Protocol (MCP) — enabling natural language querying, monitoring dashboards, and real-time data extraction.
It’s a lightweight container that connects your MCP-compatible clients (like amx-chat) to Grafana’s API.
🚀 Quick Start (via Docker)
sudo docker run -p 8010:8010 \
-e GRAFANA_API_KEY="YOUR_KEY" \
-e GRAFANA_URL="https://grafana.example.com" \
aimiox/amx-mcp-grafana:latest
🧠 You can use a Grafana service account API key with read-only access.
🔑 Learn how to create a Grafana API key using a service account
⚙️ Environment Variables
| Variable | Description |
|---|---|
GRAFANA_URL | The full URL to your Grafana instance (e.g., https://grafana.example.com) |
GRAFANA_API_KEY | A valid API key (created via a service account) |
🧠 What You Can Do
With this MCP plugin, you can:
- 🔎 Search dashboards by name
- 📋 List panels inside any dashboard
- 📊 Query panel data using natural language
- ✅ Supports Prometheus panel queries on free tier
💎 Premium Features
The free version supports querying Prometheus panels only.
The paid version adds support for:
- PostgreSQL
- Loki
- MySQL
- InfluxDB
- Elasticsearch
- ...and more!
🛠️ For paid access or custom integration, email us
🧩 Example Integration
To connect your MCP-compatible client to the Grafana MCP server, use the following configuration:
url: "http://mcp-docker-server-ip:8010/sse?amx_mcp_key=aimiox-public-mcp-key"
💡 Replace mcp-docker-server-ip with the actual IP or hostname of your MCP server container.
This configuration allows real-time interaction with your Grafana instance via MCP SSE.
💬 Sample Prompts
Here are some example natural language queries you can use with MCP clients:
- 🔍 "List all dashboards containing CPU in their name."
- 📊 "Query CPU usage panel from the Mobile Core dashboard."
🤖 Example Response (via GPT-4.1-mini)
User Prompt:
Query CPU usage panel from the Mobile Core dashboard
AI Response:
The CPU usage panel ("CPU 5-min Load") from the Mobile Core dashboard shows the 5-minute load values of the CPU over the last 6 hours, sampled every 5 minutes.
Here are some key data points from the timeframe:
- The load values fluctuate with instances of 10, 11, 24, 23, and up to 34 at some points.
- The latest data point at "2025-04-17 15:50:00" shows a load value of 23.
If you want specific value details or further analysis, please let me know!
📣 Support
For questions, issues, or contributions, open an issue on github.com/aimiox, or email us
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.