kindly-web-search-mcp-server

Overview

Kindly Web Search MCP Server is a comprehensive web search tool that provides complete content from sources like StackExchange, GitHub Issues, Wikipedia, and arXiv in optimized formats for AI assistants. Unlike basic search MCP servers, it returns the full conversation in a single call including questions, answers, comments, and metadata. The server integrates directly with APIs for specialized sources and uses a headless browser for real-time parsing of any webpage, supporting multiple search providers with intelligent fallback.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• web_searchweb_search(query, num_results=3)

  Searches the web and returns top results with title, link, snippet, and page content

  Try:Search for solutions to a JavaScript errorTry:Find recent articles about machine learningTry:Look up documentation for React hooks
• get_contentget_content(url)

  Retrieves and extracts content from a specific URL in Markdown format

  Try:Get the content from this StackOverflow answerTry:Extract the main content from this GitHub issueTry:Retrieve the full article from this blog post

Comparable tools

Installation

curl -LsSf https://astral.sh/uv/install.sh | sh

uvx --from git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server \
  kindly-web-search-mcp-server start-mcp-server

{
  "mcpServers": {
    "kindly-web-search": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server", "kindly-web-search-mcp-server", "start-mcp-server"],
      "env": {
        "SERPER_API_KEY": "your-serper-key",
        "GITHUB_TOKEN": "your-github-token"
      }
    }
  }
}

FAQ

Which search providers does Kindly support?

Kindly supports Serper (primary), Tavily, and self-hosted SearXNG with intelligent fallback between them.

Do I need a Chromium browser?

Yes, for universal HTML content extraction. However, specialized sources (StackOverflow, GitHub Issues, Wikipedia, arXiv) work well without a browser.

Compare kindly-web-search-mcp-server with