Web Researcher MCP
The AI research assistant that cites real sources honestly — and searches the web. Your AI research assistant that cites real sources and stays honest. Works with Claude, Cursor, any MCP client.
Cost / License
- Free
- Open Source (MIT)
Platforms
- Mac
- Windows
- Linux
- Online
- BSD
- Self-Hosted
- Software as a Service (SaaS)
- Docker
Features
Properties
- Lightweight
- Support for Themes
- Privacy focused
Features
- No Coding Required
- No Logs
- Extensible by Plugins/Extensions
- No registration required
- Dark Mode
- No Tracking
- Ad-free
- Full-Text Search
- Model Context Protocol (MCP) Support
- AI-Powered
- Golang
Web Researcher MCP News & Activities
Recent activities
- Maoholguin updated Web Researcher MCP
- zoharbabin rated Web Researcher MCP
- zoharbabin liked Web Researcher MCP
- zoharbabin added Web Researcher MCP
zoharbabin added Web Researcher MCP as alternative to Perplexity, Scira, Morphic and Staan
Web Researcher MCP information
Developed by
zoharbabin Licensing
Open Source (MIT) and Free product.Written in
Alternatives
9 alternatives listedSupported Languages
- English
- Zulu
- Abkhaz
- Afar
- Afrikaans
- Akan
- Albanian
- Amharic
- Arabic
- Aragonese
- Armenian
- Avaric
- Assamese
- Avestan
- Aymara
- Azerbaijani
- Bashkir
- Bambara
- Basque
- Belarusian
- Bengali
- Bihari
- Bislama
- Bosnian
- Breton
- Bulgarian
- Burmese
- Catalan; Valencian
- Chamorro
- Chechen
- Chichewa; Chewa; Nyanja
- Chinese
- Chuvash
- Cornish
- Corsican
- Cree
- Croatian
- Czech
- Danish
- Divehi; Dhivehi; Maldivian;
- Dutch
- Esperanto
- Estonian
- Ewe
- Faroese
- Fijian
- Finnish
- French
- Fula; Fulah; Pulaar; Pular
- Galician
- Georgian
- German
- Greek
- Guaraní
- Zhuang, Chuang
- Haitian; Haitian Creole
- Hausa
- Hebrew
- Herero
- Gujarati
- Hindi
- Hiri Motu
- Hungarian
- Indonesian
- Interlingua
- Interlingue
- Irish
- Igbo
- Inupiaq
- Ido
- Icelandic
- Italian
- Japanese
- Javanese
- Inuktitut
- Kalaallisut, Greenlandic
- Kannada
- Kashmiri
- Kazakh
- Kanuri
- Kikuyu, Gikuyu
- Khmer
- Kirghiz, Kyrgyz
- Komi
- Kinyarwanda
- Korean
- Kurdish
- Kongo
- Kwanyama, Kuanyama
- Latin
- Luxembourgish
- Limburgish, Limburgan, Limburger
- Lingala
- Luganda
- Lao
- Luba-Katanga
- Lithuanian
- Latvian
- Manx
- Malagasy
- Macedonian
- Malay
- Malayalam
- Maori
- Maltese
- Marathi
- Marshallese
- Nauru
- Navajo, Navaho
- Mongolian
- North Ndebele
- Norwegian Bokmål
- Nepali
- Norwegian Nynorsk
- Ndonga
- Norwegian
- Nuosu
- Ojibwe, Ojibwa
- South Ndebele
- Old Church Slavonic, Church Slavic, Church Slavonic, Old Bulgarian, Old Slavonic
- Occitan
- Oriya
- Ossetian, Ossetic
- Persian
- Panjabi, Punjabi
- Oromo
- Pashto, Pushto
- Portuguese
- Quechua
- Polish
- Yoruba
- Kirundi
- Romanian
- Pali
- Russian
- Sanskrit (Sa?sk?ta)
- Romansh
- Sindhi
- Northern Sami
- Samoan
- Sango
- Serbian
- Shona
- Sinhala, Sinhalese
- Sardinian
- Slovene
- Slovak
- Southern Sotho
- Spanish
- Somali
- Swahili
- Sundanese
- Swati
- Scottish Gaelic; Gaelic
- Tamil
- Tajik
- Telugu
- Swedish
- Tigrinya
- Tibetan Standard, Tibetan, Central
- Tagalog
- Tswana
- Tonga (Tonga Islands)
- Turkmen
- Turkish
- Tsonga
- Tatar
- Twi
- Ukrainian
- Uighur, Uyghur
- Uzbek
- Venda
- Vietnamese
- Volapük
- Walloon
- Welsh
- Wolof
- Western Frisian
- Thai
- Tahitian
- Urdu
- Yiddish
- Xhosa
AlternativeTo Categories
AI Tools & Services, Education & Reference, Online ServicesGitHub repository
- 34 Stars
- 4 Forks
- 32 Open Issues
- Updated
What is Web Researcher MCP?
web-researcher-mcp is a free, open-source MCP (Model Context Protocol) server that gives AI assistants real web search with citations that actually check out.
Instead of letting your AI hallucinate sources or pull from a closed, opaque index, you tell it exactly which sites to search — peer-reviewed journals, court databases, SEC filings, news outlets, patent offices, and more. We call these "search lenses."
Key capabilities: • Web, image, and news search across 30+ lenses (medical, legal, financial, academic, tech, and more) • Full-page extraction — reads entire articles, PDFs, Word docs, YouTube transcripts, and Hacker News threads • Academic search with real DOIs, citation graphs, and open-access links • Patent search (US, EU, international) • SEC EDGAR filing search for public-company documents • US court opinions and dockets via CourtListener • Economic data from World Bank, OECD, Eurostat, and FRED • Clinical trial search via ClinicalTrials.gov • Citation verification — checks if a source exists, matches a real record, and isn’t retracted • Bibliography auditing — validates an entire reference list in one pass • Multi-step deep research with session persistence and export • Formatted bibliographies in APA, MLA, BibTeX, RIS, and CSL-JSON
Compared to Perplexity (which gets citations wrong ~37% of the time) or ChatGPT’s built-in search, this tool verifies every link is real and every paper actually exists before citing it.
Runs on your machine — no data leaves to a third party. Free forever. Works with no API key out of the box (DuckDuckGo fallback); plug in Google, Brave, Exa, Tavily, or others for richer results.
Install in 30 seconds: uvx web-researcher-mcp or brew install zoharbabin/tap/web-researcher-mcp