What it does

When you only need to understand one paper, a full literature review is the wrong tool. Give this skill an arXiv identifier or URL and it pulls an LLM-optimised overview from AlphaXiv, falling back in tiers when a layer is missing: overview page first, then full Markdown, then the LaTeX source, so at least one layer is always readable. It draws a hard boundary, pointing you elsewhere for topic discovery and surveys while staying a quick single-paper reader. Usage is simple, run /alphaxiv with an id, with optional flags controlling how deep to read.

Works with

Claude Code; it uses curl to fetch pages from alphaxiv.org and arxiv.org and can run an external research_wiki.py against those sites.

Source and safety

It uses curl to fetch content from alphaxiv.org and arxiv.org and runs an external research_wiki.py that sends HTTP requests to both sites, flagged as justified. (This review was conducted when the skill was listed, not as ongoing monitoring. Sources are beyond our control — evaluate skills before using them.)