What problem does it solve? Gathering dozens of credible, deduplicated academic references for a manuscript is slow and error-prone. This Skill automates multi-pass scholarly retrieval, source verification, and evidence structuring so researchers get a citation-ready research packet instead of a loose list of links. ## Core Features & Use Cases - Academic evidence pipeline: Runs bounded Parallel Search passes for primary studies, reviews, seminal papers, methods, and contradictory evidence, then verifies top sources with Parallel Extract. - Manuscript research packet: Produces references in JSON and BibTeX, an evidence matrix, a claim-to-source map, synthesis of consensus and conflicts, section briefs, coverage diagnostics, and a reproducible search ledger. - Flexible backends: Defaults to Parallel Search, with explicit Parallel Research for deep reports, explicit Parallel Chat, and optional Perplexity via OpenRouter as a fallback. - Use Case: Provide a research question and context file, then receive a packet of up to 60 verified, unique references with quality labels, gaps, and Introduction/Methods/Discussion briefs. ## Quick Start Ask the AI to compile academic evidence for your manuscript research question into a research packet with verified references and an evidence matrix.
