Scholar's own interface makes date-bounded, author-bounded search a few clicks away from where you start. Here it is one call, which matters when the assistant is doing the literature sweep and you are reading the results — "papers by this author since 2020" comes back as records you can act on rather than a results page.
A small research server over Google Scholar: keyword search, filtered search, and author lookup.
- search_google_scholar_key_words — search by query string, with num_results defaulting to 5
- search_google_scholar_advanced — the same, narrowed by author and a year range
- get_author_info — details for an author by name
- Article metadata returned as structured records rather than a page of HTML
Python 3.10+, with mcp[cli], scholarly and asyncio from requirements.txt. Clone and run google_scholar_server.py, or install through Smithery for Claude, Cursor, Windsurf or Cline. The published package is mcp-server-google-scholar. It scrapes Scholar, so use it within Google's terms of service.
One command — npx -y @smithery/cli@latest install @JackKuo666/google-scholar-mcp-server --client claude