kagi-search-mcp

apridachin/kagi-search-mcp

3.3

If you are the rightful owner of kagi-search-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Kagi MCP server enables web searches using the Kagi API, providing tools to enhance model context with web content and news.

The Kagi MCP server is designed to facilitate web searches and enrich model contexts by leveraging the Kagi API. It offers a streamlined way to access and integrate web content and the latest news into model contexts, making it a valuable tool for developers and researchers who require up-to-date information and insights. The server supports several API methods, including fastgpt, enrich/web, and enrich/news, which are instrumental in providing comprehensive and enriched data. By utilizing these methods, users can enhance their applications with relevant and timely information, improving the overall performance and accuracy of their models. The server is easy to install and configure, with support for various platforms, making it accessible to a wide range of users. Additionally, the server is built with development and publishing in mind, offering straightforward processes for building, publishing, and debugging.

Features

  • Integration with Kagi API for web searches
  • Tools for enriching model context with web content
  • Support for accessing the latest news
  • Easy installation and configuration
  • Development and publishing support

Usages

installation via smithery

bash
npx -y @smithery/cli install kagi-mcp --client claude

claude desktop configuration

"mcpServers": {
  "kagi-mcp": {
    "command": "uv",
    "args": [
      "--directory",
      "path_to_project",
      "run",
      "kagi-mcp"
    ],
    "env": {
      "KAGI_API_KEY": "YOUR API KEY"
    }
  }
}

building and publishing

bash
uv sync
uv build
uv publish

debugging with inspector

bash
npx @modelcontextprotocol/inspector uv --directory path_to_project run kagi-mcp

Tools

  1. ask_fastgpt

    Search web and find an answer

  2. enrich_web

    Enrich model context with web content

  3. enrich_news

    Enrich model context with latest news