zoharbabin/web-researcher-mcp
View on GitHub ↗Give your AI assistant real web search, full-page reading, and multi-source research with citations that are never fabricated.
What you need to know
Go MCP server for citation-grade web research. 30+ research tools covering web/news/image/academic/patent/SEC/legal/econ/clinical search, full-page scraping (PDFs, Word, YouTube transcripts), citation verification and bibliography auditing, brand research, OSINT company recon, multi-step sequential research with recoverable sessions, and an LLM research panel. Search lenses restrict results to trusted sources per field, and multiple search providers can be routed with automatic failover (DuckDuckGo built-in, zero-config; Google/Brave/Serper/Exa/Tavily/SearXNG optional). Runs locally for privacy with SSRF protection, OAuth 2.1, and rate limiting in HTTP mode. Ships a Python SDK and installers for Homebrew, curl/PowerShell, Docker, Nix, and AUR. MIT.
Install
brew install zoharbabin/tap/web-researcher-mcp uvx web-researcher-mcp curl -fsSL https://raw.githubusercontent.com/zoharbabin/web-researcher-mcp/main/install.sh | sh docker.io/zoharbabin/web-researcher-mcp:latest nix run github:zoharbabin/web-researcher-mcp
Key features
- ✓30+ research tools including verify_citation, audit_bibliography, format_bibliography, and research_panel
- ✓Search lenses restrict sources to trusted domains per field
- ✓Multi-provider routing with automatic failover
- ✓Reads full articles, not just snippets
- ✓Real APA/MLA citations with working links
- ✓Guided prompt templates (comprehensive-research, fact-check, literature-review, company-recon, etc.)
- ✓Status and diagnostics resources, large-payload artifact store
Best for
Academic researchers, lawyers, journalists, medical researchers, and analysts who need verifiable citations and privacy
Reviewed 2026-08-11
Topics
- Stars
- 51★
- Forks
- 6
- Language
- Go
- License
- MIT
- Created
- 2026-05-18
- Last push
- 2026-08-17