Searches Google Scholar by keyword and scrapes back titles, authors, citation counts, and PDF links when available. Uses Chrome DevTools MCP to navigate and DOM scrape since Google Scholar has no public API. The real value is in the data-cid extraction, which is the cluster ID you need for citation exports and tracking who cited what. Includes CAPTCHA detection that pauses and asks you to solve it manually. Works best when you keep requests spaced out to avoid rate limiting. Pairs with companion skills for pagination, citation tracking, and Zotero export, all built around that data-cid identifier.
npx skills add https://github.com/cookjohn/gs-skills --skill gs-search