Metabase MCP

This is an MCP server that connects AI assistants with the Metabase analytics platform, providing functions such as data access, dashboard management, and query execution. It supports two authentication methods, facilitating direct interaction between AI and the data analytics platform.

Research and data Developer tools #Data analysis #AI integration #Metabase #MCP service .TypeScript

rating : 2 points

downloads : 7

update time : 2025-04-29

What is the Metabase MCP Server?

The Metabase MCP Server is a bridge tool used to connect AI assistants with the Metabase analytics platform. It allows AI to access resources in Metabase, such as dashboards, queries, and databases, through the intuitive `metabase://` URI.

How to use the Metabase MCP Server?

Simply configure environment variables or a JSON file to start the server and integrate it with an AI assistant. It supports multiple authentication methods and detailed logging.

Applicable scenarios

Suitable for enterprise environments that require real-time data analysis and visualization, as well as projects that aim to enhance AI interaction capabilities.

Main features

Resource accessEasily browse resources in Metabase through the `metabase://` URI.

Two authentication methodsSupports session-based login and API key authentication.

Structured data accessReturns data in JSON format, facilitating parsing and use by AI assistants.

Comprehensive loggingProvides detailed logging for easy debugging and monitoring.

Powerful error handlingClear error messages for quick problem location.

Advantages and limitations

Advantages

Seamlessly integrates AI assistants with the Metabase analytics platform.

Supports multiple authentication methods to ensure security.

Returns structured JSON data for easy parsing.

Detailed logging for easy problem troubleshooting.

Powerful error handling mechanism to enhance the user experience.

Limitations

Requires support from a Metabase instance.

Has a certain dependence on the network environment.

Advanced features may require a certain technical background for configuration.

How to use

Install dependencies

Run `npm install` to install the required Node.js dependencies.

Build the project

Use `npm run build` to build the server.

Start the server

Run `npm start` to start the server.

Usage examples

Query all dashboardsGet all dashboards in the current Metabase instance.

Execute a custom SQL queryExecute an SQL query in the specified database.

Frequently Asked Questions

How to configure API key authentication?

How to debug the MCP server?

Does it support Docker deployment?

Related resources

Official documentation

Detailed user manual and API reference.

GitHub repository

Source code and contribution guidelines.

Video tutorial

Quick start guide.

🚀 Metabase MCP Server

A Model Context Protocol (MCP) server that integrates AI assistants with the Metabase analytics platform.

🚀 Quick Start

The Metabase MCP Server serves as a bridge between your analytics platform and conversational AI, enabling seamless interaction between large language models (LLMs) with MCP capabilities and your analytical data.

✨ Features

Resource Access: Navigate Metabase resources via intuitive metabase:// URIs.
Two Authentication Methods: Supports both session-based and API key-based authentication.
Structured Data Access: Returns data in JSON format for easy consumption by AI assistants.
Comprehensive Logging: Detailed logging for debugging and monitoring.
Error Handling: Robust error handling with clear error messages.

📦 Available Tools

The server exposes the following tools for AI assistants:

Data Access Tools

list_dashboards: Retrieve all available dashboards in the Metabase instance.
list_cards: Get all saved questions/cards in Metabase.
list_databases: View all connected database sources.
list_collections: List all collections in Metabase.
list_table_columns: List the columns of a specified table.

Dashboard Operation Tools

execute_card: Execute a specified card and return the results.
get_card_result: Get the latest results of a specified card.
get_result_preview: Get a preview of the specified results.
download_result: Download the specified results as a file.
watch_cards: Subscribe to multiple cards to receive updates when changes occur.

Collection Operation Tools

list_collection_objects: List all objects (dashboards, cards, etc.) in a collection.
get_collection_object: Get a specific object in a collection.

📚 Configuration

To configure the Metabase MCP server, set the following environment variables:

API Key-based Authentication

METABASE_URL=https://your-metabase.com
METABASE_API_KEY=your_api_key

Username and Password-based Authentication

METABASE_URL=https://your-metabase.com
METABASE_USER_EMAIL=your_email@example.com
METABASE_PASSWORD=your_password

📦 Installation

To install the Metabase MCP server, follow these steps:

Clone the Repository

git clone git@github.com:hyeongjun-dev/metabase-mcp.git
cd metabase-mcp

Install Dependencies

npm install

Run the Development Server

npm run dev

🐳 Docker Support

To deploy the Metabase MCP server using Docker, execute the following commands:

Build the Docker Image

docker build -t metabase-mcp .

Run the Container

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

⚠️ Security Considerations

It is recommended to use API key authentication in a production environment.
Keep your API keys and credentials secure.
Consider using Docker secrets or environment variables instead of hard-coded credentials.
Apply appropriate network security measures to restrict access to your Metabase instance.