Keboola MCP Server

Keboola MCP Server

未知

GitHub
Created Bykeboolakeboola

CI codecov

简介

Keboola MCP 服务器允许 AI 助手（如 Claude Desktop 和 Cursor AI）与 Keboola Connection 交互，提供数据访问和分析功能。

先决条件

  • Keboola 存储 API 令牌
  • Snowflake 只读工作区凭证

安装选项

选项 1：通过 Smithery 自动安装（推荐）

npx -y @smithery/cli install keboola-mcp-server --client claude

选项 2：手动安装

git clone https://github.com/keboola/keboola-mcp-server.git
cd keboola-mcp-server
python3 -m venv .venv
source .venv/bin/activate
pip3 install -e .

配置方法

Claude Desktop 配置

  1. 编辑配置文件：

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json

  2. 添加以下配置：

{
  "mcpServers": {
    "keboola": {
      "command": "/path/to/keboola-mcp-server/.venv/bin/python",
      "args": [
        "-m",
        "keboola_mcp_server",
        "--api-url",
        "https://connection.YOUR_REGION.keboola.com"
      ],
      "env": {
        "KBC_STORAGE_TOKEN": "your-keboola-storage-token",
        "PYTHONPATH": "/path/to/keboola-mcp-server/src",
        "KBC_SNOWFLAKE_ACCOUNT": "your-snowflake-account",
        "KBC_SNOWFLAKE_USER": "your-snowflake-user",
        "KBC_SNOWFLAKE_PASSWORD": "your-snowflake-password",
        "KBC_SNOWFLAKE_WAREHOUSE": "your-snowflake-warehouse",
        "KBC_SNOWFLAKE_DATABASE": "your-snowflake-database",
        "KBC_SNOWFLAKE_SCHEMA": "your-snowflake-schema",
        "KBC_SNOWFLAKE_ROLE": "your-snowflake-role"
      }
    }
  }
}
  1. 完全退出并重新启动 Claude Desktop

Cursor AI 配置

  1. 编辑配置文件：~/.cursor/mcp.json

  2. 选择一种传输方法：

使用 SSE 传输:

{
  "mcpServers": {
    "keboola": {
      "url": "http://localhost:8000/sse?storage_token=YOUR_STORAGE_TOKEN&snowflake_account=YOUR_ACCOUNT&snowflake_user=YOUR_USER&snowflake_password=YOUR_PASSWORD&snowflake_database=YOUR_DATABASE&snowflake_schema=YOUR_SCHEMA&snowflake_warehouse=YOUR_WAREHOUSE"
    }
  }
}

使用 stdio 传输:

{
  "mcpServers": {
    "keboola": {
      "command": "/path/to/keboola-mcp-server/venv/bin/python",
      "args": [
        "-m",
        "keboola_mcp_server",
        "--transport",
        "stdio",
         "--api-url",
         "https://connection.YOUR_REGION.keboola.com"
      ],
      "env": {
        "KBC_STORAGE_TOKEN": "your-keboola-storage-token",
        "KBC_SNOWFLAKE_ACCOUNT": "your-snowflake-account",
        "KBC_SNOWFLAKE_USER": "your-snowflake-user",
        "KBC_SNOWFLAKE_PASSWORD": "your-snowflake-password",
        "KBC_SNOWFLAKE_DATABASE": "your-snowflake-database",
        "KBC_SNOWFLAKE_SCHEMA": "your-snowflake-schema",
        "KBC_SNOWFLAKE_WAREHOUSE": "your-snowflake-warehouse"
      }
    }
  }
}
  1. 重启 Cursor AI（如果使用 SSE，确保先启动服务器）

可用功能

  • 列出桶和表
  • 获取桶和表信息
  • 预览和导出表数据
  • 访问组件配置

故障排除

如果连接失败：

  • 检查 AI 助手日志
  • 验证 Keboola 令牌和 Snowflake 凭证
  • 确认所有路径为绝对路径
  • 确保虚拟环境已激活且依赖项已安装

更多信息

完整文档请访问 GitHub 仓库