mcp-databricks-server

RuniThomsen/mcp-databricks-server

3.2

If you are the rightful owner of mcp-databricks-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

MCP Databricks Server is designed to integrate Databricks with validation tools and configuration through the MCP protocol.

Tools

  1. execute_sql

    Execute SQL queries against Databricks

  2. list_tables

    List all tables in the current or specified database

  3. describe_table

    Get the schema of a specific table

MCP Databricks Server

MCP server for Databricks integration with validation tools and configuration.

Features

  • Execute SQL queries against Databricks
  • List and describe tables
  • Manage Databricks resources through MCP protocol
  • Environment-based configuration
  • Async/await support

Installation

  1. Clone this repository

  2. Install dependencies:

    pip install -r requirements.txt

Configuration

Create a .env file with your Databricks credentials:

# Your Databricks workspace URL (without https://)
DATABRICKS_HOST=your-workspace.azuredatabricks.net

# Your personal access token (generate from User Settings > Developer > Access Tokens)
DATABRICKS_TOKEN=your-access-token

# SQL warehouse HTTP path (found in SQL warehouse connection details)
DATABRICKS_HTTP_PATH=/sql/1.0/warehouses/your-warehouse-id

Usage

Run the MCP server:

python main.py

Available Tools

  • execute_sql: Execute SQL queries against Databricks
  • list_tables: List all tables in the current or specified database
  • describe_table: Get the schema of a specific table

Development

Install development dependencies:

pip install -e ".[dev]"

Run tests:

pytest

Format code:

black .

Type checking:

mypy .

License

MIT License