metabase-mcp

metabase-mcp

A TypeScript/Node.js MCP server that wraps the Metabase REST API, enabling AI agents to execute queries, explore schemas, and build dashboards through structured tool calls.

Category
Visit Server

README

Metabase MCP Server

Completely vibecoded a TypeScript/Node.js MCP (Model Context Protocol) server that wraps the Metabase REST API, giving AI agents (Claude Desktop, Claude Code) full programmatic access to a Metabase v0.59.x instance. Agents can execute queries, explore schemas, and build dashboards entirely through structured tool calls — no human click-through required.

Prerequisites

  • Node.js 20 LTS or later
  • Docker (for running a local Metabase instance during development)
  • A Metabase instance reachable over HTTP/HTTPS with an API key

Setup

  1. Install dependencies:

    npm install
    
  2. Build the server:

    npm run build
    

    Compiled output is written to dist/. The entry point is dist/index.js.

  3. Configure environment variables (copy .env.example and fill in values):

    cp .env.example .env
    
  4. (Optional) Start a local Metabase instance for development:

    docker compose up -d
    

    Metabase is available at http://localhost:3000 once healthy (allow ~2 minutes on first boot).

Environment Variables

Variable Required Description
METABASE_URL Yes Base URL of your Metabase instance, e.g. http://localhost:3000
METABASE_API_KEY Yes API key created in Admin → Settings → Authentication → API Keys

API Key Permissions: The API key must belong to a Metabase group with sufficient permissions for the operations you need. For full agent access (read schema, execute queries, create/edit cards and dashboards), the key's group should have access to all relevant databases and admin capabilities. A key with only basic viewer permissions will return 403 errors on write operations.

Adding to Claude Desktop

Add the following to your Claude Desktop MCP config file (typically ~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "metabase": {
      "command": "node",
      "args": ["/absolute/path/to/metabase-mcp/dist/index.js"],
      "env": {
        "METABASE_URL": "http://localhost:3000",
        "METABASE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Replace /absolute/path/to/metabase-mcp with the actual path to this project.

Adding to Claude Code

Add the same config to your Claude Code MCP settings:

{
  "mcpServers": {
    "metabase": {
      "command": "node",
      "args": ["/absolute/path/to/metabase-mcp/dist/index.js"],
      "env": {
        "METABASE_URL": "http://localhost:3000",
        "METABASE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Verifying the Connection

Once added to Claude Desktop or Claude Code, ask the agent to call the server_ping tool:

"Call the server_ping tool to verify the Metabase MCP server is running."

A successful response looks like:

{"ok": true, "server": "metabase-mcp", "version": "0.1.0"}

Development

# Run tests
npm test

# Watch mode
npm run test:watch

# Lint
npm run lint

# Type check without building
npm run typecheck

# Build
npm run build

# Run server directly (requires env vars set)
npm start

Architecture

  • src/client.ts — MetabaseClient class (HTTP, X-Api-Key auth, typed errors) and MetabaseApiError
  • src/index.ts — MCP server factory (createServer()), tool registrations, and stdio bootstrap
  • src/types.ts — Hand-written TypeScript interfaces for Metabase API responses

All MCP server logging goes to stderr exclusively. stdout is reserved for the JSON-RPC protocol stream — any non-protocol bytes there would corrupt the MCP session.

Metabase Version

This server targets Metabase v0.59.x. APIs introduced after this version are not used.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured