Databricks

概要

The Databricks tool provides an SDK-first Model Context Protocol (MCP) server, offering 263 robust tools and 8 prompt templates across 28 service domains. Built on the official Databricks Python SDK, it grants AI assistants deep programmatic control over Databricks resources, including Unity Catalog, SQL warehouses, Compute, Jobs, Pipelines, Serving, Vector Search, Apps, Lakebase, and more. It features flexible authentication delegation, selective tool loading for optimized performance, and built-in tool discovery, enabling seamless integration with AI clients like Claude Code, Cursor, and VS Code Copilot.

主な機能

• Comprehensive access to 263 Databricks tools across 28 service domains (e.g., Unity Catalog, SQL, Compute, Jobs, Serving)
• SDK-first design utilizing the official Databricks Python SDK for type safety and API freshness
• Zero custom authentication, leveraging Databricks SDK's automatic handling for PAT, OAuth, Azure AD, and service principals
• Selective tool loading via environment variables to optimize AI agent performance and tool selection accuracy
• Includes 8 guided MCP prompt templates for common Databricks workflows and built-in tool discovery for AI agents
• 1 GitHub stars

ユースケース

• Allow AI assistants (e.g., Claude, Cursor, Copilot) to manage Databricks resources programmatically.
• Automate complex Databricks workflows, such as deploying ML models, debugging failing jobs, or setting up data pipelines, using AI agents.
• Enable data engineers, ML engineers, and platform admins to interact with and manage their Databricks workspaces more efficiently through conversational AI.