metabase

A Model Context Protocol server that integrates AI assistants with Metabase analytics platform

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio hyeongjun-dev-metabase-mcp-server node build/index.js \
  --env METABASE_URL="https://your-metabase-instance.com" \
  --env METABASE_API_KEY="" \
  --env METABASE_PASSWORD="your_password" \
  --env METABASE_USER_EMAIL="your_email@example.com"

How to use

The Metabase MCP Server provides a bridge between Metabase analytics resources and MCP-compatible AI assistants. It exposes tools that let AI agents discover dashboards, cards, databases, and run saved questions or custom queries against your Metabase instance, returning structured JSON data suitable for downstream processing. Core tools include list_dashboards, list_cards, list_databases, execute_card, get_dashboard_cards, and execute_query. To use it, deploy the server and configure authentication (either via username/password or an API key). Once running, your MCP-enabled assistant can invoke these tools using the MCP protocol, receiving JSON-formatted responses that can drive conversational workflows and data-driven decisions. For Claude Desktop or other integrations, point the MCP client to the server’s address and provide the necessary METABASE credentials. The server also supports environment-based configuration for production deployments and can be containerized with Docker if desired.

How to install

Prerequisites:

Node.js v18.x or newer
Access to a Metabase instance with credentials or an API key
Git to clone or download the MCP server repository

Step-by-step installation:

Install dependencies

npm install

Build the project (if a build step is required by the project)

npm run build

Start the server locally

npm start

(Optional) Run in watch/auto-rebuild mode during development

npm run watch

Deploy with Docker (optional)

Build:

docker build -t metabase-mcp-server .

Run:

docker run -e METABASE_URL=https://your-metabase.com \
           -e METABASE_API_KEY=your_api_key \
           metabase-mcp-server

Configuration tips are provided in the README: set METABASE_URL and either METABASE_API_KEY or METABASE_USER_EMAIL/METABASE_PASSWORD depending on your chosen authentication method.

Additional notes

Tips and common considerations:

For production, prefer API key authentication (METABASE_API_KEY) for security.
Ensure your Metabase instance is reachable from the MCP server network, and consider restricting access to trusted clients.
Use environment variables or a .env file to manage credentials securely.
When debugging, use the MCP Inspector tool to monitor requests and responses and verify that tool calls return properly structured JSON.
If using Claude Desktop integration, update the appropriate claude_desktop_config.json to point to your server’s build/index.js path.
If you upgrade Metabase or the MCP server, re-test all tool calls to ensure compatibility with any API changes.

Related MCP Servers

zen

1.1k

Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage

MCP -Deepseek_R1

A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)

mcp-fhir

A Model Context Protocol implementation for FHIR

mcp

Inkdrop Model Context Protocol Server

mcp-appium-gestures

This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..

dubco -npm

The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.