Mcp Snowflake Server Nsp

Name: Mcp Snowflake Server Nsp
Author: nsphung

@nsphung

代码仓库

关于 Mcp Snowflake Server Nsp

A Snowflake MCP server — SQL queries, schema exploration, and data insights for AI assistants

#snowflake #mcp #key-pair #sso #oauth2

基本信息

分类

数据库

传输方式

stdio

发布者

nsphung

提交者

Nicolas PHUNG

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "snowflake": {
      "command": "uvx",
      "args": [
        "--python=3.13",
        "--from",
        "mcp-snowflake-server-nsp",
        "mcp_snowflake_server",
        "--connections-file",
        "/absolute/path/to/snowflake_connections.toml",
        "--connection-name",
        "myconn"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Mcp Snowflake Server Nsp?

A Model Context Protocol (MCP) server that connects AI assistants to Snowflake, enabling SQL queries, schema exploration, and data insights directly from an LLM client.

How to use Mcp Snowflake Server Nsp?

Configure the server using a TOML connection file to manage multiple environments (e.g., production, staging), choose an authentication method (password, key-pair, external browser, or TOML file), and invoke it with optional flags such as --exclude-json-results, --exclude_tools, or --prefetch to pre-load schemas.

Key features of Mcp Snowflake Server Nsp

Multiple authentication methods (password, key-pair, external browser, TOML)
TOML multi-connection config for different environments
Write-safety guard (disabled by default, must be explicitly enabled)
Exclusion patterns to filter databases, schemas, or tables
--exclude-json-results flag to reduce context window usage
Prefetch mode to pre-load table schemas as MCP resources

Use cases of Mcp Snowflake Server Nsp

Run SQL queries on Snowflake from an AI assistant (e.g., Claude Desktop)
Explore Snowflake schemas and table structures without leaving your chat
Derive data insights by combining natural language with query results
Manage multiple Snowflake environments (production, staging, development) via one config file
Pre-load table schemas to speed up repeated schema lookups