General-Purpose Snowflake

General-Purpose Snowflake MCP server for any type of SQL operation

python

GitHub Stars

python

Language

4 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "sfc-gh-tjia-snowflake-generic-mcp": {
      "command": "/opt/homebrew/bin/uv",
      "args": [
        "--directory",
        "/path/to/your/project",
        "run",
        "snowflake_mcp_server.py"
      ],
      "env": {
        "MCP_LOG_LEVEL": "DEBUG",
        "SNOWFLAKE_ROLE": "YOUR_ROLE",
        "SNOWFLAKE_SCHEMA": "YOUR_SCHEMA",
        "SNOWFLAKE_ACCOUNT": "YOUR_SNOWFLAKE_ACCOUNT",
        "SNOWFLAKE_DATABASE": "YOUR_DATABASE",
        "SNOWFLAKE_MAX_ROWS": "100",
        "SNOWFLAKE_PASSWORD": "YOUR_PASSWORD",
        "SNOWFLAKE_USERNAME": "YOUR_USERNAME",
        "SNOWFLAKE_WAREHOUSE": "YOUR_WAREHOUSE",
        "SNOWFLAKE_AUTHENTICATOR": "externalbrowser",
        "SNOWFLAKE_PRIVATE_KEY_PATH": "/path/to/rsa_key.p8",
        "SNOWFLAKE_PRIVATE_KEY_PASSPHRASE": "YOUR_PASSPHRASE"
      }
    }
  }
}

You can run a secure MCP server that lets AI assistants access your Snowflake data warehouse through natural language interactions. It translates your natural language requests into SQL, executes them safely, and returns formatted results with helpful error messages and usage guidance.

How to use

Connect your MCP client to the Snowflake MCP server to begin interacting with your data. You can ask questions in natural language, such as requesting table schemas, listing available tables, filtering data, or analyzing trends. The server executes SQL queries against Snowflake and returns results in nicely formatted tables with metadata and row counts. It also provides detailed error messages and troubleshooting tips if something goes wrong.

How to install

# Prerequisites: Python and a virtual environment
# 1) Clone the project
git clone https://github.com/sfc-gh-tjia/snowflake-generic-mcp.git
cd snowflake-generic-mcp

# 2) Create and activate a virtual environment
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# 3) Install dependencies
pip install -r requirements.txt

Configuration basics

Set up your Snowflake credentials and optional settings in environment variables. At minimum you need your account identifier and username. You can enable password, private key, or SSO authentication. Optional settings configure the warehouse, database, schema, role, and maximum rows returned per query.

Available tools

query_execution

Executes SQL queries against Snowflake and returns results with formatting and metadata.

error_handling

Provides detailed, actionable error messages and troubleshooting guidance.

security_logging

Logs dangerous operations and enforces security policies during query execution.

performance_limits

Applies configurable limits to protect against resource exhaustion.

nl_to_sql

Converts natural language prompts into SQL queries via the MCP protocol integration.

Built by

VeilStrat

AI signals for GTM teams