Metabase

Metabase MCP Server connects your Metabase instance to AI assistants through the Model Context Protocol (MCP), enabling natural language access to dashboards, charts, and BI assets. You can ask questions, generate dashboards from descriptions, and manage access and connections through simple MCP client configurations.

How to use

Use the Metabase MCP Server to connect a Metabase instance with any MCP client. Open your MCP client and add the server definition using either a local stdio connection or a remote HTTP connection. Once connected, you can ask questions about dashboards and charts, generate new visuals by describing what you want, and perform common BI actions like listing databases, creating cards, or updating dashboards through natural language prompts.

Practical usage patterns include: creating a new dashboard by describing the metrics you want, querying existing dashboards or cards, listing users or groups, and modifying BI assets through conversational commands. If you run into issues, check the connection settings, validate your Metabase URL and API key, and ensure the MCP transport is accessible from your MCP client.

How to install

Prerequisites: you need a Metabase instance running, Python available, and the uv package manager installed.

Step 1: Install uv Package Manager on your system.

For Windows you can run:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

For macOS or Linux you can run:

curl -LsSf https://astral.sh/uv/install.sh | sh

Alternatively, install via package managers if you prefer: macOS: brew install uv ; Windows via Scoop: scoop install uv ; Windows via Chocolatey: choco install uv.

# macOS
brew install uv

# Windows (via Scoop)
scoop install uv

# Windows (via Chocolatey)
choco install uv

Step 2: Acquire the MCP server code

Clone the project repository or download the development bundle to your local machine.

# Example clone
git clone https://github.com/codewalnut/metabase-mcp-server.git
cd metabase-mcp-server

Step 3: Install dependencies

Install dependencies and set up a virtual environment using the project’s tooling.

uv sync

Step 4: Configure your Metabase credentials

Configure how the server authenticates with Metabase. You can use a .env file, command-line arguments, or environment variables in your MCP client configuration.

Example using a .env file in the project root:

METABASE_URL=http://localhost:3000
METABASE_API_KEY=mb_xxx_your_key
PORT=3200
HOST=localhost
TRANSPORT=streamable-http
LOG_LEVEL=DEBUG

Step 5: Connect to your MCP client

Choose an MCP client (for example Claude Desktop, Cursor, Windsurf, or others) and add the Metabase MCP server using either a stdio (local) or streamable-http (remote) transport.

stdio transport (local, recommended for development):

{
  "mcpServers": {
    "metabase": {
      "type": "stdio",
      "command": "C:\\Users\\YourName\\Projects\\metabase-mcp-server\\.venv\\Scripts\\python.exe",
      "args": [
        "C:\\Users\\YourName\\Projects\\metabase-mcp-server\\src\\metabase_mcp_server.py"
      ],
      "env": {
        "METABASE_URL": "http://localhost:3000",
        "METABASE_API_KEY": "mb_xxx_your_key",
        "PORT": 3200,
        "HOST": "localhost",
        "TRANSPORT": "streamable-http",
        "LOG_LEVEL": "DEBUG"
      }
    }
  }
}

Step 6: Start the server and test

Run the start command for the local stdio setup and verify the MCP server launches successfully.

If you choose the HTTP transport, you can connect with this remote URL:

{
  "mcpServers": {
    "metabase": {
      "type": "streamable-http",
      "url": "http://localhost:3200/mcp/"
    }
  }
}

Available tools

get_metabase_collection

Get a collection by ID

create_metabase_collection

Create a new collection

update_metabase_collection

Update collection metadata

delete_metabase_collection

Delete a collection

get_metabase_cards

List all charts

get_card_query_results

Get results from a chart query

create_metabase_card

Create a new chart

update_metabase_card

Update an existing chart

delete_metabase_card

Delete a chart

get_metabase_dashboards

List dashboards

get_dashboard_by_id

Get a dashboard by ID

get_dashboard_cards

Get cards in a dashboard

get_dashboard_items

Get all dashboard items

create_metabase_dashboard

Create a dashboard

update_metabase_dashboard

Update a dashboard

delete_metabase_dashboard

Delete a dashboard

copy_metabase_dashboard

Create a copy of an existing dashboard

get_metabase_databases

List databases

create_metabase_database

Create a new database connection

update_metabase_database

Update a database connection

delete_metabase_database

Delete a database connection

get_metabase_users

List all users

get_metabase_current_user

Get current user details

create_metabase_user

Create a new user

update_metabase_user

Update user info

delete_metabase_user

Delete a user

get_metabase_groups

List user groups

create_metabase_group

Create a user group

delete_metabase_group

Delete a user group

execute_sql_query

Execute a native SQL query