What problem does it solve? Finding high-quality scientific and technical information on the web is difficult with generic search engines that surface blogs and SEO content over primary research. This Skill provides semantic web search and batch URL content extraction through the Exa API, with built-in levers to bias results toward research papers and academic domains. ## Core Features & Use Cases - Semantic Web Search: Run natural-language queries via the Exa Python SDK with search types (auto, fast, deep), date filters, domain allowlists/blocklists, and a research-paper category for scholarly retrieval. - Batch URL Extraction: Fetch full text or highlight snippets from multiple URLs in a single call, including academic PDFs and arXiv pages, with structured metadata (title, author, published date). - Academic Source Strategy: A two-pass pattern combining a scholarly-filtered search with a general search, plus citation formatting rules (author-year, DOI links, mandatory Sources section). - Use Case: A researcher asks for recent findings on CRISPR off-target effects. The Skill runs an academic search restricted to arxiv.org, pubmed.ncbi.nlm.nih.gov, and nature.com, extracts the top papers, and returns a synthesized answer with inline citations and a grouped Sources list. ## Quick Start Set your EXA_API_KEY environment variable, then ask the assistant to search the web for a scientific topic using Exa with the research paper category and save the results to a JSON file.
