Databricks

关于

This template empowers users to create a Model Context Protocol (MCP) server that runs on Databricks Apps, forming a vital bridge between AI assistants like Claude and your Databricks workspace. It allows for seamless integration by letting you define what an AI can see and do within Databricks, through easily configurable markdown-based prompts and custom Python functions that leverage the Databricks SDK. The server dynamically loads prompts, exposes functions as MCP tools, and manages communication via HTTP requests and Server-Sent Events, handling authentication automatically.

主要功能

• Add AI prompts using simple markdown files for dynamic loading.
• Provides an OAuth-enabled proxy for seamless authentication with Databricks.
• Exposes Python functions as MCP tools via a dedicated decorator.
• Create custom AI tools with Python functions leveraging the Databricks SDK.
• 0 GitHub stars
• Handles both HTTP requests and MCP protocol over Server-Sent Events (SSE).

使用案例

• Facilitating AI-driven creation and management of Databricks jobs or other workspace resources.
• Enabling AI assistants to execute SQL queries directly against Databricks SQL Warehouses.
• Allowing AI assistants to query Databricks for system information, cluster status, or file listings.