A web search and content extraction skill built on Exa's API, optimized for scientific and technical queries. The main draw is the research paper filtering: you can pass category flags and domain allowlists to bias results toward arXiv, PubMed, Nature, and similar scholarly sources instead of generic SEO content. It wraps two Exa endpoints in Python CLI scripts, one for semantic search and one for batch URL extraction including PDFs. The routing guide in the source is clear about when to use search versus extract, and the academic source priority section gives you concrete domain lists and a two pass strategy for finding peer reviewed work. Useful if you're building research agents or just tired of wading through blog spam when you want actual papers.
npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill exa-search