lalig7

Zotero Ref Import V2 — Development skill for Claude Code

Development community

Claude skill that imports references into a Zotero collection from Word bibliographies, reference lists, single PDFs or PDF folders.

How to install Zotero Ref Import V2

This entry records only its repository, not the path inside it, so there is no exact command to give. Open lalig7/Zotero_Ref_import_V2 and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Zotero Ref Import V2 does

Claude skill that imports references into a Zotero collection from Word bibliographies, reference lists, single PDFs or PDF folders. Matches each item to a verified PMID or DOI via PubMed and CrossRef, checks titles to catch wrong DOIs, skips duplicates, and handles Zotero rate limits. Your API key stays in your own terminal.

Alternatives in Development

Learn From Materials — Turn PDFs, books and papers into interactive learning webpages｜将复杂材料转化为可追溯、可测验、可做笔记的学习网页 225 ★
Pubmed MCP Server — Search PubMed/Europe PMC, fetch articles and full text (PMC/EPMC/Unpaywall), citations, MeSH terms via MCP 150 ★
Nature Academic Search — Academic Paper Search：面向中文科研用户的 Codex / Claude Code 文献检索 Skill + MCP；跨 CrossRef、PubMed、arXiv、OpenAlex、Europe P 132 ★

README

Zotero Reference Import: an AI agent skill

Get your references into Zotero without typing them in one by one. This skill works in **Claude** (Claude desktop, Claude Code, Claude.ai) and **OpenAI Codex**. It imports references into the Zotero collection you choose, from:

You have…	What happens
A Word bibliography (`.docx`)	Each reference is resolved by PMID → DOI → URL and created as a full Zotero record (authors, journal, year, volume, pages)
A plain-text reference list (`.txt` / `.md`), e.g. a literature-search shortlist	Same as Word, one reference per line
A single research PDF	A PubMed- or CrossRef-verified record is created and the PDF is attached
A folder of research PDFs (subfolders included)	Either every PDF is uploaded and attached to its record, or (faster, for citations only) the PDFs are turned into a verified reference list that is imported like a Word bibliography

Metadata comes from **PubMed** and **CrossRef**. Each match is checked against the paper's title before anything is created, so a typo or a wrong DOI printed in a PDF can't silently import the wrong paper.

Safety by design

Preview first. Nothing is written to Zotero unless you add --upload.
Your API key is never stored. It is typed at a terminal prompt (or read from ZOTERO_API_KEY) and is never written to scripts, logs or progress files.
Resumable. Progress is saved after every item, so you can re-run the same command after an interruption. References already in the collection (matched by PMID, DOI or URL) are skipped, not duplicated.
Gentle on Zotero. Uploads pause between batches and wait out Zotero's rate limit instead of failing.
Nothing is deleted or merged. Mismatches and references with no identifier are listed for you to review by hand.