databricks-mcp-server-by-cdata

databricks-mcp-server-by-cdata

0
databases

CData's MCP Server for Databricks facilitates live data querying by LLMs using natural language, wrapping the CData JDBC Driver. This read-only server supports comprehensive data access and enables seamless interaction with Databricks data.

databricks-mcp-server-by-cdata

CData's Model Context Protocol (MCP) Server for Databricks is a read-only MCP server that allows LLMs like Claude Desktop to query live Databricks data. Powered by the CData JDBC Driver, this server makes Databricks data accessible through a simple MCP interface, enabling natural language querying without SQL. It supports full read, write, update, delete, and action capabilities via a simplified setup.

Setup Guide

  • Clone and build the server from the repository.
  • Download and install the CData JDBC Driver for Databricks.
  • Configure your connection and create a JDBC connection .prp file.

Usage Details

Once configured, AI clients can interact with the database through natural language questions.

Tools & Descriptions

  • get_tables - Retrieves a list of tables.
  • get_columns - Retrieves columns for a specified table.
  • run_query - Executes SQL SELECT queries.

Troubleshooting

  • Ensure Claude Desktop is fully quit and reopened.
  • Verify connection configurations.
  • Contact CData Support for assistance.

License

This project is open under the MIT License.