zotero-mcp connects a Zotero research library to AI assistants through the Model Context Protocol, allowing them to search papers, retrieve content, summarize research, and work with citations or annotations. It is for people who want to discuss and analyze their research library with an AI assistant. The catalogue skill and MCP entry provide that connection.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add 54yyyu/zotero-mcp --skill zotero-cligit clone --depth 1 https://github.com/54yyyu/zotero-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/54yyyu/zotero-mcp/zotero-cli)<a href="https://agentmods.dev/skills/54yyyu/zotero-mcp/zotero-cli"><img src="https://agentmods.dev/badge/skills/54yyyu/zotero-mcp/zotero-cli/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/54yyyu/zotero-mcp/zotero-cli"><img src="https://agentmods.dev/badge/skills/54yyyu/zotero-mcp/zotero-cli.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00093 | $0.01508 |
| Opus 5 | $0.00046 | $0.00754 |
| Sonnet 5 | $0.00019 | $0.00302 |
| Haiku 4.5 | $0.00009 | $0.00151 |
Grade A, and why
zotero-cli scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Zotero from the shell
zotero-cli reaches a Zotero library directly. Prefer it over the Zotero MCP
server when you have shell access: the MCP server's tool schemas cost ~14k
tokens of context on every request whether or not you use them, while this
costs only what you actually run.
Check it is set up before the first real call:
zotero-cli config # prints the resolved Zotero settings
If that fails, Zotero is not reachable. Local mode needs the Zotero desktop
app running with its local API enabled; web mode needs ZOTERO_API_KEY and
ZOTERO_LIBRARY_ID. Say so rather than guessing at library contents.
Always pass --json when you will read the output
Default output is markdown for humans. --json gives you one object per
invocation with a stable shape, so you never parse prose:
{"ok": true, "command": "search", "schema": 1, "data": {...}}
{"ok": false, "command": "search", "schema": 1, "error": {"message": "...", "code": "..."}}
Both go to stdout; [INFO]/[WARN] diagnostics go to stderr. Check ok
before using data. Run zotero-cli --json-schema for the full contract.
The core loop
Almost every task is: find keys → act on keys. Item keys are 8 characters and are the currency of every command.
# 1. find - use --detail keys_only to keep the result small while browsing
zotero-cli --json search "attention mechanisms" --limit 10 --detail keys_only
# 2. act - metadata, full text, or a page range
zotero-cli --json get metadata ABCD1234
zotero-cli --json get fulltext ABCD1234
zotero-cli --json read ABCD1234 --start-page 3 --end-page 8
Pipe keys straight into the next call:
zotero-cli --json search "diffusion models" --limit 5 --detail keys_only \
| jq -r '.data.items[].key' \
| while read -r key; do zotero-cli --json get metadata "$key"; done
Choosing a search mode
| Mode | Use it for | Command |
|---|---|---|
items (default) |
a title, author or phrase you know | search "Vaswani attention" |
semantic |
a topic or idea, no exact wording | search --mode semantic "why transformers scale" |
tag |
items you filed under a tag | search --mode tag "to-read,important" |
advanced |
structured field conditions | search --mode advanced --conditions '[...]' |
citekey |
a BibTeX citation key | search --mode citekey smith2020 |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 145 lines · 93 tokens per session scan A 4e8177843b3b
zotero-cli is a skill published in the GitHub repository 54yyyu/zotero-mcp (4,981 stars, last pushed 16d ago), licensed MIT. It adds 93 tokens to every session and 1,508 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
zotkit
Use when the user wants to add, create, upload/attach, download/fetch, search, tag, look up by item key, export BibTeX for, or organize references (papers, PDFs) in their Zotero library via the zotkit CLI — headless Zotero management through the Web API + WebDAV, no desktop app. Covers task recipes and safety rules.…
zotio
Use when the user wants to search, script, or automate a Zotero library — even if they don't say "Zotero" or "BibTeX": local-first library search, CI-gateable health reports, preview-first writes with undo, reviewable PDF/DOI import, annotation export, Obsidian/Logseq vault sync, plus an MCP server for agents. Trigger…
paper-fulltext-harvest
Batch download academic paper full-text (PDF/XML) from a list of DOIs. Handles 25 DOI prefixes across 19 publisher families via three layered routes: (1) publisher TDM APIs requiring institutional subscription (Elsevier ScienceDirect, Wiley Online, Springer Nature), (2) Open Access sources (Crossref, Unpaywall…
zotero-connector
Import arXiv papers into Zotero with duplicate detection and batch support. TRIGGER when: user asks to import papers, add arxiv papers to zotero, batch import papers, check for duplicate papers in zotero, mentions importing by arXiv ID, or wants to list Zotero collections. DO NOT TRIGGER when: user is working on…
citation-management
Comprehensive citation management for academic research. Search Google Scholar and PubMed for papers, extract accurate metadata, validate citations, and generate properly formatted BibTeX entries. This skill should be used when you need to find papers, verify citation information, convert DOIs to BibTeX, or ensure…
liteparse
Local document and PDF parsing with spatial text and bounding boxes. Use for extracting text from PDFs, DOCX, Office files, and images; OCR on scans; layout-preserved JSON for RAG; batch-ingesting paper folders; or page screenshots for multimodal agents — even when the user does not name liteparse. Prefer over…