Skill Detail

Ground research agents in Zotero libraries with LLM for Zotero

Bring agent-mode paper chat, citation-grounded answers, note saving, tagging, importing, and library-wide research operations into Zotero.

Research & ScrapingMulti-Framework
Research & Scraping Multi-Framework Security Reviewed
⭐ 1.7k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill ground-research-agents-in-zotero-libraries-with-llm-for-zotero Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Zotero 7 or newer, the LLM for Zotero XPI plugin, an LLM backend or local/OpenAI-compatible model, optional Codex App Server or Claude Code bridge
Install & setup
Download the latest XPI from the upstream GitHub releases page, install it in Zotero through Tools > Add-ons > Install Add-on From File, restart Zotero, then configure the preferred model provider and optional agent-mode bridge in the plugin preferences.
Author
yilewang
Publisher
Individual
Last updated
May 28, 2026
Quick brief

Use LLM for Zotero when an agent needs to work inside a researcher’s actual Zotero library instead of treating papers as loose uploaded PDFs. The operator invokes it from Zotero to chat with the current paper, compare open papers, ask library-wide questions, save grounded notes, tag or import items, and route longer research work through Codex App Server or Claude Code where configured. The boundary is Zotero-centered literature workflow and library operations; it is not a generic chatbot, citation manager listing, or standalone MCP server card.

How it works

What this skill actually does

Inputs and prerequisites: Zotero 7 or newer, the LLM for Zotero XPI plugin, an LLM backend or local/OpenAI-compatible model, optional Codex App Server or Claude Code bridge.

Setup notes: Download the latest XPI from the upstream GitHub releases page, install it in Zotero through Tools > Add-ons > Install Add-on From File, restart Zotero, then configure the preferred model provider and optional agent-mode bridge in the plugin preferences.

Source and verification boundary: use https://yilewang.github.io/llm-for-zotero as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.

Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.