zoharbabin/web-researcher-mcp repository overview

⁠web-researcher-mcp

An MCP (Model Context Protocol) server providing web search, content extraction, and multi-source research tools for AI assistants.

⁠Features

• Web Search — Multi-provider support (Google, Brave, Serper, SearXNG) with search lenses
• Content Extraction — 4-tier scraping pipeline: Markdown → Stealth HTTP → Standard HTTP → Headless Browser
• YouTube Support — 3-strategy transcript extraction with description fallback
• Academic Search — Scholar, arXiv, and PubMed integration
• Patent Search — Multi-office filtering (US, EP, WO, CN, JP, KR)
• News & Image Search — With freshness and size filters
• Sequential Research — Multi-step research workflows with context tracking

⁠Quick Start

⁠Docker

docker pull zoharbabin/web-researcher-mcp:latest

docker run -e GOOGLE_CUSTOM_SEARCH_API_KEY=your-key \
 -e GOOGLE_CUSTOM_SEARCH_ID=your-id \
 zoharbabin/web-researcher-mcp:latest

⁠Binary

Download from GitHub Releases⁠.

⁠Configuration

⁠Architecture

Built in Go with zero global state, interface-driven design, bounded concurrency, and defense-in-depth security (SSRF protection, rate limiting, content sanitization).

Multi-arch images: linux/amd64 and linux/arm64.

⁠Links

• GitHub Repository⁠
• Documentation⁠
• License: MIT⁠

Tag summary