ai-skill-arxiv

Fetch and analyze an arXiv paper via tiered fallback (markdown -> arXiv HTML -> ar5iv -> PDF). Rate-limited for arxiv2md (28 req/min, deterministic). Produces a structured summary with citation, problem, key claims, method, results, limitations. Use when the user says "analyze arxiv paper", "summarize arxiv", "read paper", "what does this paper say", or provides an arXiv ID/URL to analyze.

Monitor arXiv for new papers matching saved queries/categories. Manages a watchlist with deterministic "seen" tracking so subsequent checks return only new material. Use when the user says "watch arxiv", "track arxiv", "new papers on", "arxiv monitor", "check my arxiv watches", or wants periodic discovery. Depends on arxiv-search (must be installed as a sibling skill).

Search arXiv for research papers by topic, category, author, or date range. Returns structured JSON with titles, authors, abstracts, and links - no full text (pair with arxiv-analyze for that). Use when the user says "search arxiv", "find arxiv papers", "arxiv papers on", "latest papers about", or "arxiv research".