The Snowflake Model Context Protocol (MCP) server provides a comprehensive toolkit for AI agents to interact with Snowflake's robust data platform. It seamlessly integrates advanced Snowflake Cortex AI capabilities—including Cortex Search for unstructured data, Cortex Analyst for rich semantic modeling, and Cortex Agent for orchestrating data retrieval—with essential data management functions like object creation, modification, and SQL execution. By connecting to various MCP clients such as Claude Desktop, Cursor, fast-agent, and GitHub Copilot, users can empower their AI assistants to intelligently query, manage, and analyze both structured and unstructured data within Snowflake, all while respecting fine-grained SQL statement permissions and role-based access control.
主な機能
01Object Management for basic Snowflake object operations (create, drop, update)
02SQL Execution with user-configured permissions for LLM-generated queries
03Cortex Analyst for querying structured data via semantic modeling
04Cortex Search for querying unstructured data in RAG applications
05189 GitHub stars
06Cortex Agent for orchestrating structured and unstructured data retrieval
ユースケース
01Building Retrieval Augmented Generation (RAG) applications using Snowflake data
02Creating natural language data assistants integrated with IDEs like VS Code
03Empowering AI agent clients (e.g., Claude, Cursor, fast-agent) to interact with Snowflake's data platform
