OSS Snowflake MCP Server
by snowflake-labscommunityv1.3.5
272
86
The Snowflake MCP server provides tools for Snowflake Cortex AI, enabling object management and SQL orchestration within the MCP ecosystem. It supports various capabilities including querying unstructured and structured data, executing SQL, and managing Snowflake objects.
From the registry: MCP Server for Snowflake from Snowflake Labs
Install & configure
Prerequisites
- requiredSnowflake accountActive Snowflake account with appropriate role permissions
- requiredService config fileCreate a tools_config.yaml defining Cortex services and permissions
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `mcp-snowflakelabs` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Snowflake account** — Active Snowflake account with appropriate role permissions
- **Service config file** — Create a tools_config.yaml defining Cortex services and permissions
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["snowflake-labs-mcp","--service-config-file","<path_to_file>/tools_config.yaml","--connection-name","default"]
- optional environment variables:
- `SNOWFLAKE_ACCOUNT`: Snowflake account identifier (example: `<your-account>`)
- `SNOWFLAKE_USER`: Snowflake username (example: `<your-username>`)
- `SNOWFLAKE_PASSWORD`: Snowflake password or PAT (example: `<your-password>`)
- `SNOWFLAKE_ROLE`: Role to use
- `SNOWFLAKE_WAREHOUSE`: Warehouse to use
- `SNOWFLAKE_PRIVATE_KEY`: Private key for key-pair auth
- `SNOWFLAKE_PRIVATE_KEY_FILE`: Path to private key file
- `SNOWFLAKE_PRIVATE_KEY_FILE_PWD`: Private key file password
Note: Prefers connections.toml via --connection-name. Also supports streamable-http for container deployments on port 9000.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: Prefers connections.toml via --connection-name. Also supports streamable-http for container deployments on port 9000.
Tools (6)
Cortex Search
Query unstructured data in Snowflake for Retrieval Augmented Generation applications.
Cortex Analyst
Query structured data in Snowflake via rich semantic modeling.
Cortex Agent
Agentic orchestrator for structured and unstructured data retrieval.
Object Management
Perform basic operations against Snowflake's common objects.
SQL Execution
Run LLM-generated SQL managed by user-configured permissions.
Semantic View Querying
Discover and query Snowflake Semantic Views.
Environment variables
SNOWFLAKE_ACCOUNTrequiredAccount identifier for Snowflake.SNOWFLAKE_USERrequiredUsername for authentication.SNOWFLAKE_PASSWORDrequiredPassword or programmatic access token.SNOWFLAKE_ROLERole to use for connection.SNOWFLAKE_WAREHOUSEWarehouse to use for queries.Packages
snowflake-labs-mcp
pypiv1.3.5transport: stdio
Details
- Language
- Python
- License
- Apache-2.0
- Released
- Oct 28, 2025
- Last commit
- Apr 2, 2026
- Registry name
- io.github.Snowflake-Labs/mcp