@rog0x/MCP Web Tools

@rog0x

リポジトリ

基本情報

カテゴリ

検索

ライセンス

MIT

作者

rog0x

@rog0x/MCP Web Tools について

MCP server for AI agents: web content extraction, DuckDuckGo search, site monitoring, HTML to markdown. No API keys needed. Install: npx @rog0x/mcp-web-tools

@rog0x/MCP Web Tools の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "web-tools": {
      "command": "npx",
      "args": [
        "@rog0x/mcp-web-tools"
      ]
    }
  }
}

@rog0x/MCP Web Tools のツール

Extract structured content from a URL: title, headings, paragraphs, links, images, and metadata

Search the web using DuckDuckGo and return structured results with titles, URLs, and snippets

Check if a website is up, measure response time, and get HTTP details

Convert HTML content or a URL to clean, readable markdown. Removes scripts, styles, and navigation.

Extract content from multiple URLs in parallel. Returns results for each URL.

概要

What is @rog0x/mcp-web-tools?

A Model Context Protocol server that provides AI agents with web capabilities: content extraction, web search via DuckDuckGo, site monitoring, and HTML‑to‑markdown conversion, all through a clean tool interface.

How to use @rog0x/mcp-web-tools?

Install by adding a mcpServers entry to your Claude Code or Claude Desktop config using npx @rog0x/mcp-web-tools, or globally via npm install -g @rog0x/mcp-web-tools and run mcp-web-tools. Invoke the five exposed tools: web_extract, web_search, web_monitor, html_to_markdown, and web_multi_extract.

Key features of @rog0x/mcp-web-tools

No API keys required — uses DuckDuckGo for search.
Fast parallel extraction with 15‑second request timeout.
Clean output — strips scripts, styles, and ads.
Lightweight — minimal dependencies, no headless browser.
Production‑ready error handling and input validation.

Use cases of @rog0x/mcp-web-tools

Extract structured content (titles, headings, links) from any URL.
Search the web using DuckDuckGo without an API key.
Monitor website uptime, response time, and HTTP headers.
Convert any URL or HTML into clean Markdown.
Batch extract content from multiple URLs in parallel.

よくある質問

What runtime does @rog0x/mcp-web-tools require?

Node.js 18 or newer.

Do I need an API key for web search or content extraction?

No — web search uses DuckDuckGo and all other tools work over direct HTTP without any API keys.

How does the server handle timeouts and errors?

Each request has a 15‑second timeout, and the server includes proper error handling and input validation.

Is a headless browser needed to extract content?

No — the server is lightweight and works without a headless browser, using direct HTTP requests.

Under what license is @rog0x/mcp-web-tools distributed?

MIT — rog0x.

「検索」の他のコンテンツ

Default Privacy

Sourced privacy answers for AI assistants: how private an app or service is, private replacements for mainstream apps, privacy tool facts with sources, guides, MCP server vetting and free domain checks. Read-only, no acc

Singapore Proxy

Singapore Mobile Proxy

Give Claude, Cursor or any MCP agent a real Singapore mobile IP. Fetch pages, run Singapore-localised Google searches and check geo-locked content through physical Singtel and M1 SIMs, with IP rotation on demand. Hosted

Anymail Finder

Find and verify B2B work emails from Claude, ChatGPT, Cursor and other AI assistants. Find a person's email from their name and company or LinkedIn URL, find a decision maker by role, list the emails at a company, verify

Desearch

Desearch-ai

AI search, X search and web search for AI agents, plus page extraction and X data tools. Bring your own Desearch API key.

OpenCityRecord

OpenCity

Search city council meetings, votes, and documents across U.S. cities, with citations to the official record.