Fetches web content as clean Markdown with site handlers, caching, search, extraction, and fallback fetch strategies.
Key Features: Multi-stage web fetching with archive and metadata fallbacks Dedicated handlers for YouTube, arXiv, Wikipedia, GitHub, and Twitter/X Search with Brave, SearXNG, or DuckDuckGo fallbacks Open-source MCP server for web search, content extraction, and library docs indexing with hybrid search and built-in embeddings.
Key Features: Embedded SearXNG-based web search Content extraction with multi-step research Hybrid search combining full-text (FTS5) and semantic embeddings Searches the web through multiple free engines and fetches content from web pages, GitHub READMEs, and CSDN.
Key Features: Multi-engine web search without API keys Structured titles, URLs, and descriptions Web, Markdown, GitHub README, and CSDN fetching Searches, scrapes, crawls, maps, and researches web content through Firecrawl-compatible MCP tools.
Key Features: Web, news, image, GitHub, and developer search URL scraping with structured output options Bounded multi-page crawling and site mapping Searches DuckDuckGo and fetches readable webpage content for MCP clients.
Key Features: Webpage content fetching and parsing Regional search selection Tavily MCP — wraps the Tavily API (tavily.com)
Searches, scrapes, and extracts live public web data through MCP, with browser automation and structured results.
Key Features: Google, Bing, and Yandex web search Markdown and HTML page scraping Structured extraction for major web platforms MCP server enabling LLMs to explore and interact read-only or optionally write on Fediverse via ActivityPub protocol.
Key Features: WebFinger discovery and actor fetching Timeline fetching and thread retrieval Instance exploration and trending content Fetches web pages, APIs, articles, and YouTube transcripts as HTML, JSON, text, or Markdown through MCP.
Key Features: HTML, Markdown, plain-text, and JSON fetching Mozilla Readability article extraction YouTube transcript retrieval with language selection AI-powered MCP server providing real-time web search and training-data knowledge via OpenRouter models.
Key Features: Supports multiple OpenRouter AI models including Perplexity Sonar and Grok 4 Real-time web search with configurable timeouts and model tiers Structured citation extraction from AI responses Fetches JavaScript-rendered web pages with Playwright and returns extracted Markdown or HTML, including parallel multi-URL retrieval.
Key Features: Playwright-based JavaScript rendering Readability content extraction Local-first MCP web research with keyless search, browsing, crawling, extraction, caching, and similarity discovery.
Key Features: Headless browser fetching Multi-mode website crawling