Search and analyze Zotero research libraries through MCP
Use Zotero MCP to let research agents search papers, retrieve metadata, inspect annotations, and analyze citation context from a Zotero library.
npx skills add agentskillexchange/skills --skill search-and-analyze-zotero-research-libraries-through-mcp
Use Zotero MCP when a research or writing agent needs source-backed access to a Zotero library instead of relying on pasted citations, browser search, or model memory. The operator installs zotero-mcp-server, connects a local or web Zotero library, optionally builds semantic or full-text indexes, and lets an MCP-capable client search papers, retrieve metadata, inspect notes and PDF annotations, analyze citations, or prepare literature-review context. Invoke it when the agent’s task depends on the user’s curated research corpus: paper discovery, annotated reading summaries, related-work drafting, citation checks, or evidence collection across a library. The scope boundary is Zotero library retrieval and analysis through MCP and its companion CLI; it is not a general research assistant, citation manager listing, PDF parser, or broad knowledge-management product card.
What this skill actually does
Inputs and prerequisites: Zotero, Zotero local API or web API credentials, Python package tooling such as uv, pip, or pipx, and an MCP-capable client.
Setup notes: Install the base server with uv tool install zotero-mcp-server, pip install zotero-mcp-server, or pipx install zotero-mcp-server, then run zotero-mcp setup to configure supported MCP clients. Add optional extras such as zotero-mcp-server[semantic], zotero-mcp-server[pdf], zotero-mcp-server[scite], or zotero-mcp-server[all] when semantic search, PDF extraction, or citation intelligence is needed.
Source and verification boundary: use https://stevenyuyy.com/zotero-mcp/ 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 MCP workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.