Snowflake MCP - Data Warehouse Queries for AI Agents¶
What It Is¶
Snowflake MCP exposes Snowflake data warehouse operations over the Model Context Protocol. AI agents can run SQL queries, explore database schemas, manage virtual warehouses, and analyze enterprise data - all RBAC-aware with the user's own Snowflake permissions.
Tools Available¶
| Tool | Description |
|---|---|
| SQL query execution | Run read-only or read-write queries through the agent |
| Schema exploration | List databases, schemas, tables, views, and columns |
| Warehouse management | Start, stop, resize, and monitor virtual warehouses |
| Query history | Review past queries, execution times, and credits consumed |
Quick Start¶
npx -y mcp-snowflake-server
Business Use Cases¶
- Ad-hoc analysis: "What were our top 10 customers by revenue last quarter?"
- Schema discovery: "Show me all tables in the sales schema with column definitions"
- Cost management: "Which queries consumed the most credits this week?"
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.
Limitations¶
- Snowflake account required. RBAC applies - agent can only access what the authenticated user can. Write operations need explicit enablement.