What problem does it solve? Finding the right research papers for a query is hard: a single keyword search misses sibling methods, related work, and papers whose abstracts don't literally mention the topic. This Skill turns Firecrawl's research paper index — covering PubMed, bioRxiv, medRxiv, and arXiv — into a structured literature-finding workflow that matches the search strategy to the type of question being asked. ## Core Features & Use Cases - Semantic paper search: Query paper abstracts with HyDE-based semantic search via the firecrawl_research_search_papers MCP tool or firecrawl research search-papers CLI command. - Citation and similarity expansion: Turn one good hit into a full relevant set using related_papers with similar, citers, or references modes. - In-body verification: Confirm load-bearing claims (methods used, scores reported, affiliations) with read_paper, and fetch canonical metadata with inspect_paper. - Use Case: Ask "what are alternatives to the Adam optimizer?" and the Skill runs an initial semantic search, expands strong anchors through citation graphs, and returns the full family of relevant papers ranked by relevance. ## Quick Start Ask the agent to find papers on your research question, for example: find papers about training-free detection of AI-generated text and include closely related methods.
