Official Snowflake MCP server for data warehouse
## Snowflake MCP Server: Cloud Data Warehouse The **Snowflake MCP Server** integrates Snowflake's cloud data platform into Google Antigravity, enabling developers to query, manage, and analyze massive datasets through AI-assisted SQL development and data engineering workflows. ### Why Snowflake MCP? - **Elastic Scaling**: Scale compute independently from storage for optimal cost/performance - **Multi-Cloud**: Run on AWS, Azure, or GCP with consistent experience - **Zero-Copy Cloning**: Instantly clone databases for development and testing - **Time Travel**: Query historical data and recover from accidental changes - **Data Sharing**: Securely share live data across organizations without copying ### Key Features #### 1. SQL Queries and Data Analysis ```sql -- Execute queries via MCP with AI optimization suggestions SELECT DATE_TRUNC('month', order_date) as month, product_category, SUM(revenue) as total_revenue, COUNT(DISTINCT customer_id) as unique_customers FROM analytics.orders WHERE order_date >= DATEADD(year, -1, CURRENT_DATE()) GROUP BY 1, 2 ORDER BY 1 DESC, 3 DESC; ``` #### 2. Data Engineering Operations ```python # Manage warehouses and resources await snowflake.executeCommand( "ALTER WAREHOUSE analytics_wh SET WAREHOUSE_SIZE = LARGE" ) # Clone database for development await snowflake.executeCommand( "CREATE DATABASE dev_analytics CLONE production_analytics" ) # Time travel query historical_data = await snowflake.query( "SELECT * FROM orders AT(OFFSET => -3600)", warehouse="analytics_wh" ) ``` ### Configuration ```json { "mcpServers": { "snowflake": { "command": "npx", "args": ["-y", "@anthropic/mcp-snowflake"], "env": { "SNOWFLAKE_ACCOUNT": "your-account", "SNOWFLAKE_USER": "your-username", "SNOWFLAKE_PASSWORD": "your-password", "SNOWFLAKE_WAREHOUSE": "default_warehouse", "SNOWFLAKE_DATABASE": "default_database" } } } } ``` ### Use Cases **Analytics Engineering**: Build and maintain data transformation pipelines with dbt-style development workflows. **Ad-Hoc Analysis**: Run complex analytical queries on massive datasets with instant scalability. **Data Platform Development**: Manage Snowflake resources, permissions, and data sharing configurations from your IDE. The Snowflake MCP Server brings enterprise data warehouse capabilities to AI-assisted development.
{
"mcpServers": {
"snowflake": {
"mcpServers": {
"snowflake": {
"env": {
"SNOWFLAKE_USER": "YOUR_USER",
"SNOWFLAKE_ACCOUNT": "YOUR_ACCOUNT",
"SNOWFLAKE_PASSWORD": "YOUR_PASSWORD",
"SNOWFLAKE_WAREHOUSE": "YOUR_WAREHOUSE"
},
"args": [
"-y",
"@snowflake-labs/mcp-server-snowflake"
],
"command": "npx"
}
}
}
}
}