lightdash-mcp-server

What is lightdash-mcp-server?

lightdash-mcp-server is an MCP (Model Context Protocol) server that provides AI assistants with standardized access to the Lightdash API. It allows users to explore and retrieve data from their Lightdash projects, including projects, spaces, charts, dashboards, and metrics.

How to use lightdash-mcp-server?

Install the server via npm (npm install lightdash-mcp-server), configure a .env file with your Lightdash API key and API URL, then start the server with npx lightdash-mcp-server. The server exposes tools that can be called by any MCP-compatible AI client.

Key features of lightdash-mcp-server

• List all projects in your Lightdash organization
• Retrieve details of a specific project
• List spaces, charts, and dashboards within a project
• Get custom metrics and catalog information
• Export charts and dashboards as code

Use cases of lightdash-mcp-server

• An AI assistant listing all available projects before querying specific data
• Automatically fetching chart definitions or dashboard layouts for analysis
• Retrieving metric catalogs to understand available dimensions and measures
• Exporting Lightdash content as code for version control or reuse

FAQ from lightdash-mcp-server

What runtime does lightdash-mcp-server require?

It requires Node.js and npm. The server is run via npx or as a local install.

How do I authenticate with Lightdash?

Set the environment variables LIGHTDASH_API_KEY and LIGHTDASH_API_URL (default https://app.lightdash.cloud/api/v1) in a .env file.

Does the server store any data?

No—it acts as a proxy, making requests to the Lightdash API on behalf of the AI assistant. No data is stored locally.

What transport does lightdash-mcp-server use?

It uses the standard MCP transport (stdio) for communication with the AI client.

What are the dependencies?

The server depends on the Lightdash API being accessible and the user providing a valid API key. No external database or other services are required.