mcp-snowflake-server

This repository provides a Model Context Protocol (MCP) server for interacting with Snowflake databases, offering tools for querying, schema exploration, and appending data insights to a dynamic memo resource.

57
6

Snowflake MCP Server

smithery badge PyPI - Version

Overview

This is a Model Context Protocol (MCP) server designed to facilitate interaction with Snowflake databases. It allows users to execute SQL queries and interact with a dynamic "insights" memo.

Components

Resources

  • memo://insights: A resource that stores and dynamically updates data insights discovered during analysis, automatically updating via the append_insight tool.

Tools

The server provides six core tools:

Query Tools

  • read_query: Executes SELECT queries, returning results as an array of objects.
  • write_query: Executes INSERT, UPDATE, or DELETE queries (requires --allow-write flag), returning the number of affected rows.
  • create_table: Creates new tables (requires --allow-write flag), returning confirmation.

Schema Tools

  • list_tables: Lists all tables in the database, returning an array of table names.
  • describe-table: Provides column information for a specified table, returning an array of column definitions.

Analysis Tools

  • append_insight: Adds data insights to the memo://insights resource, triggering an update.

Usage with Claude Desktop

The server can be installed via Smithery, UVX, or locally, requiring configuration in claude_desktop_config.json with Snowflake connection details. Options include enabling write access, setting log directories/levels, and excluding specific tools.

Repository

IS
isaacwasserman

isaacwasserman/mcp-snowflake-server

Created

December 13, 2024

Updated

March 28, 2025

Language

Python

Category

Database