Redash

This MCP server provides access to Redash data sources and query execution through a lightweight, configurable bridge. You can run queries, wait for results, and retrieve data sources and their details from your MCP-enabled client or automation workflow.

How to use

You interact with the Redash MCP server through your MCP client by selecting the redash MCP server you configured and invoking the available actions. The core capabilities are to execute a SQL query against a data source, wait for the results to become available, list all data sources, and fetch details about a specific data source. Use these tools to integrate Redash data into your workflows, dashboards, or automation routines.

How to install

Prerequisites you need before running the MCP server are Node.js for development and Docker if you choose the container route. Prepare API access to Redash by obtaining an API key and knowing your Redash base URL.

# Prerequisites
node -v
npm -v

# Optional for container workflow
docker -v

# Environment values you will set later
# REDASH_API_KEY=your_api_key
# REDASH_BASE_URL=https://redash.example.com

Configuration and running modes

Two practical ways to run the Redash MCP server are using npx for a quick, system-wide package execution or running a Docker container for isolated deployment. You can also run in SSE mode for real-time messaging if you want to expose the server as an SSE endpoint during development or after building.

{
  "mcpServers": {
    "redash": {
      "command": "npx",
      "args": [
        "mcp-redash"
      ],
      "env": {
        "REDASH_API_KEY": "<YOUR_API_KEY>",
        "REDASH_BASE_URL": "https://redash.example.com"
      }
    }
  }
}

Additional run options and examples

If you prefer running via Docker, use the container image and pass the necessary environment variables. The following configuration shows how to wire the Docker-based MCP server into your client setup.

{
  "mcpServers": {
    "redash_docker": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e", "REDASH_API_KEY",
        "-e", "REDASH_BASE_URL",
        "yuki9541134/mcp-redash"
      ],
      "env": {
        "REDASH_API_KEY": "<YOUR_API_KEY>",
        "REDASH_BASE_URL": "https://redash.example.com"
      }
    }
  }
}

Security and environment considerations

Keep your Redash API key secret. Do not commit the key to public repositories. Use environment variables to inject the key at runtime and restrict access to the MCP server process. When deploying in production, prefer a secure secret management solution and rotate keys regularly.

Notes on server modes

In SSE mode you can start the HTTP server and communicate via Server-Sent Events. Development runs with npm and the dev server flag, while a built version can be started with Node. This mode exposes endpoints for streaming messages and accepting input.

Tools and endpoints

The Redash MCP server exposes the following actions you can call through your MCP client: execute_query_and_wait to run a query and wait for results, list_data_sources to enumerate data sources, and get_data_source to fetch details of a particular data source.

Available tools

execute_query_and_wait

Executes a SQL query against a specified data source and waits until the results are ready, returning the query result.

list_data_sources

Returns a list of all available Redash data sources.

get_data_source

Fetches detailed information for a specific data source by its ID.