Snowflake MCP Server

What does “Snowflake” MCP Server do?

The Snowflake MCP Server is an implementation of the Model Context Protocol (MCP) that connects AI assistants and developer tools to a Snowflake database. It enables seamless database interaction by allowing users to execute SQL queries, manage database schemas, and access data insights through standardized MCP interfaces. By exposing Snowflake’s data and schema as accessible resources and providing tools for reading, writing, and managing tables, the server empowers AI-powered workflows, agents, and LLMs to perform database tasks. This dramatically enhances developer productivity by automating data analysis, table management, and schema exploration, all within secure and configurable boundaries.

List of Prompts

No prompt templates are explicitly mentioned in the repository or documentation.

List of Resources

• memo://insights
  • A continuously updated memo that aggregates discovered data insights. It is updated automatically when new insights are appended via the append_insight tool.
• context://table/{table_name}
  • (Available if prefetch is enabled) Provides per-table schema summaries, including columns and comments, exposed as individual resources.

List of Tools

• read_query
  • Executes SELECT SQL queries to read data from the Snowflake database, returning results as an array of objects.
• write_query (enabled only with --allow-write)
  • Executes INSERT, UPDATE, or DELETE SQL modification queries, returning the number of affected rows or a confirmation message.
• create_table (enabled only with --allow-write)
  • Allows creation of new tables in the Snowflake database using a CREATE TABLE SQL statement and returns a confirmation of table creation.
• list_databases
  • Lists all databases in the Snowflake instance, returning an array of database names.
• list_schemas
  • Lists all schemas within a specified database.
• list_tables
  • Lists all tables within a specific database and schema, returning table metadata.
• describe_table
  • Provides column information for a specific table, including names, types, nullability, defaults, and comments.

Use Cases of this MCP Server

• Database Management and Exploration
  • Developers and AI agents can automate the process of listing, describing, and managing databases, schemas, and tables within Snowflake, streamlining data infrastructure management.
• Automated Data Analysis
  • Run parameterized queries to extract insights, generate reports, or feed downstream analytics pipelines.
• Schema Discovery and Documentation
  • Automatically fetch and summarize schema details for documentation, compliance, or onboarding new team members.
• Contextual Data Insights
  • Use the memo://insights resource to aggregate and access evolving data insights, supporting collaborative analytics or audit trails.
• Table Creation and Data Engineering
  • Programmatically create tables and update data via secure, auditable write operations, enabling automated ETL, data ingestion, or transformation workflows.

How to set it up

Windsurf

1. Ensure you have Node.js installed and access to your Windsurf configuration.
2. Open your Windsurf configuration file (often windsurf.json).
3. Add the Snowflake MCP Server as a new entry in the mcpServers array:

   {
     "mcpServers": [
       {
         "command": "mcp-snowflake-server",
         "args": ["--port", "8080"]
       }
     ]
   }
   

4. Save the configuration and restart Windsurf.
5. Verify connection to the Snowflake MCP Server in the Windsurf interface.

Securing API Keys (Example)

{
  "command": "mcp-snowflake-server",
  "env": {
    "SNOWFLAKE_ACCOUNT": "your_account",
    "SNOWFLAKE_USER": "your_user",
    "SNOWFLAKE_PASSWORD": "${SNOWFLAKE_PASSWORD}"
  },
  "inputs": {
    "database": "your_db"
  }
}

Claude

1. Ensure Claude supports MCP server integrations.
2. Locate your Claude configuration file or MCP integration settings.
3. Add the Snowflake MCP Server as a source:

   {
     "mcpServers": [
       {
         "command": "mcp-snowflake-server",
         "args": []
       }
     ]
   }
   

4. Save changes and restart Claude.
5. Confirm that Claude recognizes and can interact with the Snowflake MCP Server.

Cursor

1. Install required dependencies and access Cursor’s configuration.
2. Open the cursor.json or equivalent settings file.
3. Insert the Snowflake MCP Server in the mcpServers block:

   {
     "mcpServers": [
       {
         "command": "mcp-snowflake-server",
         "args": []
       }
     ]
   }
   

4. Save and restart Cursor.
5. Check Cursor’s status page for MCP server connectivity.

Cline

1. Make sure Cline is installed and up-to-date.
2. Open the Cline configuration file.
3. Register the Snowflake MCP Server as follows:

   {
     "mcpServers": [
       {
         "command": "mcp-snowflake-server",
         "args": []
       }
     ]
   }
   

4. Save the configuration and restart Cline.
5. Validate the connection to the Snowflake MCP Server.

Note on Securing API Keys

Store sensitive credentials such as Snowflake passwords or API tokens using environment variables. Reference them securely in your configuration files using the env property.

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "snowflake-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “snowflake-mcp” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Based on the above, the Snowflake MCP Server offers a robust set of tools and resources for Snowflake database interaction, but lacks prompt templates and explicit sampling/roots support information.

Our opinion

The Snowflake MCP Server provides comprehensive Snowflake database access tools and useful resource primitives, is well-documented, and includes practical security/configuration guidance. However, the absence of prompt templates and explicit roots/sampling support reduces its MCP completeness. Overall, it is a strong and practical MCP implementation for database workflows.

MCP Score