DEV Community

curatedmcp for CuratedMCP

Posted on • Originally published at curatedmcp.com

Snowflake MCP: Let Your AI Agent Query Your Data Warehouse Directly

Install guide and config at curatedmcp.com

Snowflake MCP: Let Your AI Agent Query Your Data Warehouse Directly

The Snowflake MCP server is the official bridge between AI agents and your Snowflake data warehouse. Instead of copying data into prompts or building custom APIs, Claude, Cursor, and other AI tools can now execute SQL queries, inspect schemas, manage compute, and surface analytics insights directly—with proper authentication and access controls.

What It Does

Snowflake MCP gives AI agents native capabilities to work with enterprise data:

  • Query execution: Run SQL and get structured results back
  • Schema inspection: Explore tables, views, and column definitions programmatically
  • Warehouse management: Scale compute up or down, monitor usage
  • Performance visibility: Access query history and execution metrics
  • Advanced data features: Create tasks, manage streams, query external tables and data shares
  • BI integration: Surface insights that feed into analytics workflows

This unlocks practical workflows: an AI agent can investigate a slow dashboard query, suggest optimizations, or pull analysis for a report—all without leaving your chat interface. For analytics teams, it means treating your data warehouse like a first-class citizen in your AI tooling, not an afterthought.

How to Install

Install the server in your environment:

npx -y @snowflake/mcp
Enter fullscreen mode Exit fullscreen mode

Then add it to your Claude Desktop config:

{
  "mcpServers": {
    "snowflake-mcp": {
      "command": "npx -y @snowflake/mcp",
      "env": {
        "SNOWFLAKE_ACCOUNT": "your-account-id",
        "SNOWFLAKE_USER": "your-username",
        "SNOWFLAKE_PASSWORD": "your-password",
        "SNOWFLAKE_WAREHOUSE": "your-warehouse",
        "SNOWFLAKE_DATABASE": "your-database"
      }
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Real-World Use Cases

  • Debugging dashboards: Ask Claude to analyze a slow dashboard query, review execution plans, and suggest index or materialization strategies—all without manual SQL inspection.
  • Ad-hoc analytics: Generate exploratory queries on the fly. "What are our top 10 customers by revenue this quarter?" Claude writes and executes the query in real time.
  • Warehouse optimization: Monitor unused tables, check compute costs, and identify queries to refactor—let your AI agent surface cost-saving opportunities automatically.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Top comments (0)