What problem does it solve? Writing a literature review requires searching multiple academic databases, verifying citation authenticity, and synthesizing sources into coherent arguments. This Skill automates multi-database literature search, enforces strict no-fabrication rules for citations, and structures sources into evidence-claim mappings before drafting review sections. ## Core Features & Use Cases - Multi-Database Search: Query PubMed, CrossRef, Semantic Scholar, and arXiv via the scholar_search.py script with year filtering and result limits. - Citation Format Export: Output results as BibTeX, RIS, APA, MLA, Chicago, or Vancouver formats for direct use in LaTeX or reference managers. - Evidence-Claim Mapping: Convert retrieved papers into a structured table linking each source to a supported claim and its citation slot in the manuscript. - Chinese Literature Workflow: Generate CNKI search strategies and keywords, then organize user-provided abstracts into the review structure. - Use Case: A graduate student writing a related work section searches "federated learning intrusion detection" across CrossRef and Semantic Scholar, exports BibTeX entries, builds an evidence-claim map, and drafts a synthesized review paragraph with verified citations. ## Quick Start Ask the assistant to search academic databases for papers on your research topic and organize the results into a literature review outline with verified citations.
