Databricks MCP Server

This is a server that provides tools to interact with Databricks.

Features

• Cluster management

• Job management

• Notebook management

• DBFS management

• SQL management

Installation

Required Python version: 3.11

Configuration

The server can be configured using environment variables or a .env file.

Environment variables

• DATABRICKS_HOST: The Databricks host URL (e.g. https://adb-123456789012345.12.azuredatabricks.net)

• DATABRICKS_TOKEN: The Databricks API token

• TRANSPORT: The transport to use (e.g. sse). Options: [stdio, sse, streamable-http]. Default: sse

• SERVER_HOST: The host to bind the server to (e.g. 0.0.0.0)

• SERVER_PORT: The port to bind the server to (e.g. 8000)

• DEBUG: Whether to run the server in debug mode (e.g. True)

• LOG_LEVEL: The log level (e.g. INFO)

.env file

Create a .env file in the root directory of the project with the following variables:

Obtaining Databricks Credentials

1. Host: Your Databricks instance URL (e.g., your-instance.cloud.databricks.com)

2. Token: Create a personal access token in Databricks:

   • Go to User Settings (click your username in the top right)

   • Select "Developer" tab

   • Click "Manage" under "Access tokens"

   • Generate a new token, and save it (it will not be shown again)

Running the server in standalone mode

Config Claude-Desktop/Cursor/Windsurf

Add the following to your Claude-Desktop/Cursor config file:

If you are using Claude-Desktop, restart it for the changes to take effect.