mcp_server_for_databricks

mcp_server_for_databricks

3.2

If you are the rightful owner of mcp_server_for_databricks 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 Server for Databricks Interaction provides an interface to a Databricks workspace using the FastMCP framework, enabling users to query and retrieve information about Databricks resources through MCP commands.

The MCP Server for Databricks Interaction is designed to simplify metadata retrieval tasks for users interacting with Databricks via the MCP interface. It leverages the Databricks SDK and CLI for backend communication, providing tools that enhance the effectiveness of agents in writing code for Databricks workspaces. The server uses OAuth U2M authentication to connect to Databricks, eliminating the need for token creation and storage. It provides metadata for catalog objects and sample data for tables, enriching the context for users. The server is built on the FastMCP framework and interacts with Databricks using the databricks-sdk for Python, handling authentication and information retrieval efficiently.

Features

  • OAuth U2M authentication for secure connection to Databricks workspace.
  • Provides metadata for catalogs, schemas, and tables using Databricks SDK.
  • Retrieves sample data for tables to enrich metadata context.
  • Handles authentication and information retrieval using databricks-cli and SDK.
  • Exposes tools for querying Databricks resources via MCP commands.