MCP Snowflake Service

🚀 Model Context Protocol (MCP) Server - Claude Snowflake Database Access Service

This server implements the Model Context Protocol (MCP), enabling Claude to:

• Execute SQL queries on the Snowflake database.
• Automatically manage the database connection lifecycle (connection creation, timeout reconnection, and connection closure).
• Handle query results and errors.
• Perform database operations securely.

🚀 Quick Start

📦 Installation

Install via Smithery

To automatically install mcp-service-snowflake for Claude Desktop via Smithery:

npx -y @smithery/cli install @datawiz168/mcp-service-snowflake --client claude

Manual Installation

If you need to manually run the server for testing, you can use the following command:

python server.py

⚠️ Important Note

You don't need to manually start the server during normal use. The Claude Desktop client will automatically manage the server's startup and shutdown according to the configuration.

⚙️ Configuration Instructions

Environment Variable Configuration

Add the following content to ~/.config/claude/settings.json:

{
  "snowflakeEnabled": true,
  "snowflakeEndpoint": "your_snowflake_endpoint",
  "snowflakeUser": "your_snowflake_user",
  "snowflakePassword": "your_snowflake_password"
}

✨ Features

• Secure Snowflake Database Access: Precise control of access permissions through database users.
• Robust Error Handling and Reporting Mechanism: Ensure timely detection and handling of errors.
• Automatic Connection Management: Automatically establish connections when the server starts and release resources when it stops.
• Query Execution and Result Handling: Support the execution of complex SQL queries and efficient result handling.

💻 Usage Examples

Basic Usage

1. Install Dependencies Install the required Python packages:

   pip install snowflake-sqlalchemy sqlalchemy click
   

2. Configure Snowflake Access Permissions For security reasons, it is recommended to create a database user with only read permissions and use it for this service.
3. Start the Server Use the following command to start the MCP server:

   python server.py
   

🔧 Technical Details

Contribute Code or Report Issues

1. Fork this repository.
2. Create a 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. Submit a Pull Request.

Note

The MCP-Server-Snowflake precisely controls access permissions through database users. If you only need to read data, provide the user with read-only database permissions.

📄 License

This project is licensed under the MIT License.