JordiNei/mcp-databricks-server
Free连接到 Databricks API，允许 LLM 运行 SQL 查询、列出作业和获取作业状态。
FreeFree tier
About JordiNei/mcp-databricks-server
A Model Context Protocol (MCP) server that connects to Databricks API, enabling LLMs to execute SQL queries on Databricks SQL warehouses, list all jobs, and retrieve job status and details. Requires Python 3.7+ and a Databricks workspace with a personal access token and SQL warehouse endpoint. Supports natural language interaction with Databricks when used with MCP-compatible LLMs.
Key Features
Run SQL queries on Databricks SQL warehouses
List all Databricks jobs in workspace
Get status of specific Databricks jobs by ID
Get detailed information about Databricks jobs
Supports natural language interaction with LLMs via MCP protocol
Pros & Cons
Pros
- Open source and free to use
- Simple setup with environment variables for credentials
- Specific integration with Databricks SQL and Jobs API
- Compatible with any MCP-supporting LLM
Cons
- Requires a Databricks workspace with personal access token and SQL warehouse
- Only supports SQL queries and job operations (no notebooks, MLflow, or other Databricks features)
- Security risk if token or .env file is exposed
Best For
Query Databricks tables and databases using natural languageMonitor and check the status of Databricks jobsRetrieve detailed job information for debugging or reportingIntegrate Databricks operations into AI assistant workflows
FAQ
What are the prerequisites for using this MCP server?
Python 3.7+, a Databricks workspace with a personal access token, a SQL warehouse endpoint, and permissions to run queries and access jobs.
How do I set up the server?
Clone the repository, create a virtual environment, install dependencies (pip install -r requirements.txt), create a .env file with DATABRICKS_HOST, DATABRICKS_TOKEN, and DATABRICKS_HTTP_PATH, then run python main.py.
What MCP tools does it provide?
The server provides four tools: run_sql_query (execute SQL), list_jobs (list all jobs), get_job_status (status by job ID), and get_job_details (detailed info by job ID).
How can I test the connection?
Run the included test script: python test_connection.py