muthuramanathanm/mcp-snowflake-service

🚀 Enhanced MCP Snowflake Server v2.0

A powerful Model Context Protocol (MCP) server that provides Claude with comprehensive Snowflake database access capabilities.

✨ Enhanced Features

This server provides 7 powerful tools for comprehensive Snowflake exploration:

🔍 Discovery Tools

• list_databases - Show all available databases
• list_schemas - List schemas in databases
• list_tables - Show tables in schemas
• search_tables - Find tables by pattern

📊 Analysis Tools

• describe_table - Show detailed table structure
• get_table_sample - Retrieve sample data
• execute_query - Run custom SQL queries

🐳 Production Ready

• Docker containerization with health checks
• Comprehensive documentation (3 reference formats)
• Production-grade deployment setup
• 50+ ready-to-use prompts

🚀 Quick Start

Option 1: Docker Deployment (Recommended)

1. Clone the repository

git clone https://github.com/muthuramanathanm/mcp-snowflake-service.git
cd mcp-snowflake-service

2. Configure credentials

# Create .env file with your Snowflake credentials
cp .env.example .env
# Edit .env with your actual credentials

3. Build and run with Docker

docker build -t mcp-snowflake-server .
# The Docker wrapper script handles everything automatically

Option 2: Local Installation

1. Clone and install dependencies

git clone https://github.com/muthuramanathanm/mcp-snowflake-service.git
cd mcp-snowflake-service
pip install -r requirements.txt

⚙️ Configuration

Cursor IDE MCP Configuration

Add the following to your Cursor MCP configuration file (~/.cursor/mcp.json):

Docker Deployment (Recommended)

{
  "mcpServers": {
    "snowflake": {
      "mode": "stdio",
      "command": "/path/to/mcp-snowflake-service/docker-mcp-wrapper.sh",
      "cwd": "/path/to/mcp-snowflake-service"
    }
  }
}

Local Python Deployment

{
  "mcpServers": {
    "snowflake": {
      "mode": "stdio", 
      "command": "python",
      "args": ["server.py"],
      "cwd": "/path/to/mcp-snowflake-service"
    }
  }
}

Configuration Notes:

• Docker approach: Uses the included wrapper script for containerized deployment
• Local approach: Runs Python directly (requires local Python environment)
• Update paths: Replace /path/to/mcp-snowflake-service with your actual installation path

Snowflake Credentials

Create a .env file in the project root directory:

SNOWFLAKE_USER=your_username
SNOWFLAKE_PASSWORD=your_password  
SNOWFLAKE_ACCOUNT=your_account.region    # Example: autodesk-stg.us-east-1
SNOWFLAKE_DATABASE=your_database
SNOWFLAKE_WAREHOUSE=your_warehouse

Security Note: Never commit the .env file to version control. Add it to your .gitignore.

🔄 Connection Management

The server provides robust automatic connection management:

• Automatic initialization

  • Creates connection when first query is received
  • Validates connection parameters and credentials
  • Handles authentication and session setup

• Connection maintenance

  • Tracks connection state and health
  • Handles connection timeouts gracefully
  • Automatically reconnects if connection is lost
  • Implements retry logic with exponential backoff

• Resource management

  • Properly closes connections when server stops
  • Releases database resources appropriately
  • Prevents connection leaks and orphaned sessions

📚 Usage & Examples

Natural Language Prompts

The server supports intuitive natural language queries. Here are some examples:

Discovery Commands

• "What databases are available?"
• "What schemas are in EDH_PUBLISH database?"
• "Find tables with 'subscription' in the name"
• "What tables are in the PARTNER_WEB_SERVICES_PRIVATE schema?"

Analysis Commands

• "Show me the structure of the ACCOUNT table"
• "Give me 5 sample rows from the SUBSCRIPTION table"
• "Describe the CONTRACTLINEITEM_INT table"

Custom SQL

• "Execute: SELECT COUNT() FROM EDH_PUBLISH.EDH_SHARED.SUBSCRIPTION"*
• "Run: SELECT * FROM ACCOUNT WHERE ID IS NOT NULL LIMIT 3"

Manual Testing

For development and testing, you can run the server standalone:

# Local testing
python server.py

# Docker testing  
./test-docker-mcp.sh

🌟 Key Features

• 🔍 7 Powerful Tools - Complete Snowflake exploration toolkit
• 🐳 Docker Ready - Production-grade containerized deployment
• 📖 Rich Documentation - 3 comprehensive reference formats
• 🚀 Natural Language - Intuitive prompts for immediate productivity
• 🔒 Secure - Robust error handling and credential management
• ⚡ High Performance - Optimized connection pooling and caching

📁 Documentation

This repository includes comprehensive documentation:

• MCP_PROMPTS_REFERENCE.md - Complete guide with 50+ examples and workflows
• QUICK_PROMPTS.txt - Copy-paste ready prompts for immediate use
• PROMPTS_REFERENCE.json - Structured reference for automation
• README_DOCKER.md - Detailed Docker deployment guide

🛠️ Development

To contribute or report issues:

1. Fork this repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

🔐 Security Notes

• Access Control: Snowflake access is controlled through database user permissions
• Read-Only Use: For read-only access, use a Snowflake user with read-only permissions
• Credential Security: Never commit .env files or credentials to version control
• Network Security: Use appropriate firewall rules and VPN access as needed

📄 License

This project is licensed under the .

Built with ❤️ for the developer community. Star ⭐ this repo if you find it useful!