What problem does it solve? Finding the right research papers for a query is hard: a single keyword search misses relevant work, and general web search returns pages rather than paper records. This Skill queries Firecrawl's research paper index — a corpus dominated by PubMed, bioRxiv, and medRxiv biomedical literature plus arXiv preprints in CS, physics, and math — to return the full relevant set of papers for any research question. ## Core Features & Use Cases - Semantic paper search: Use firecrawl_research_search_papers for HyDE-based semantic search over paper abstracts as the first move for almost any query. - Semantic and structural expansion: Use firecrawl_research_related_papers with modes similar, citers, and references to turn one good hit into a complete set of neighboring methods and related work. - Metadata and in-body verification: Use firecrawl_research_inspect_paper for canonical citation metadata and firecrawl_research_read_paper to verify load-bearing claims inside a paper's full text. - Use Case: A user asks "what are alternatives to the Adam optimizer?" The Skill runs an initial semantic search, expands several strong anchor papers with related_papers in similar mode, re-seeds from new hits, and returns the full family of relevant optimizer papers ranked by relevance. ## Quick Start Ask the assistant to find papers answering your research question, for example: "Find papers on training-free detection of AI-generated text and include closely related methods."
