researchzy-downloadlisted
Install: claude install-skill Researchzy/Researchzy-academic-skills
# Researchzy Download
This skill retrieves full text through lawful open-access sources and the publisher APIs the user is entitled to. `scripts/batch_download.mjs` is the entry point; metadata, open-access resolution, publisher APIs, content validation and the manifest are separate modules under `scripts/lib/`. Resolve `SKILL_DIR` to the directory containing this `SKILL.md` and run the bundled scripts by that path, never relative to the working directory.
It covers no institutional or browser access: it does not sign in to libraries, reuse browser sessions or pass verification pages. When a paper has no open-access copy and no API the user holds a key for, report it and let the user supply the file.
## Before downloading
1. Establish a definite list of papers: DOIs, exact titles, or legitimate PDF URLs. Do not turn a keyword search into bulk downloading, and do not download whole journal issues, volumes or result sets. Process only the papers the user identified or confirmed.
2. Researchzy first. When the Researchzy tools are available, look each paper up with `search_papers` (pass its DOI as `doi`, an arXiv ID as `arxiv`, a title as `query`) and call `get_pdf_download` for it. When that returns a URL, download and verify the file as `researchzy` describes and skip the downloader for that paper. Run the downloader for the papers Researchzy has no PDF for, or for all of them when the tools are not available.
3. Prefer DOIs. Resolve a title to its DOI with Researchzy searc