54yyyu/zotero-mcp

Zotero MCP: Connects your Zotero research library with Claude and other AI assistants via the Model Context Protocol to discuss papers, get summaries, analyze citations, and more.

What it solves

Zotero MCP allows users to integrate their Zotero research library directly into AI assistants like Claude and ChatGPT. It eliminates the need to manually upload papers or copy-paste text by giving AI agents the ability to search, read, and organize a user's academic library.

How it works

The project implements the Model Context Protocol (MCP), providing a standardized way for AI clients to interact with Zotero. It can operate in local mode (reading directly from the zotero.sqlite database) or via the web API. It offers two primary integration paths: an MCP server for chat-based AI apps and a zotero-cli skill for coding agents with shell access.

Who it’s for

Researchers, academics, and students who use Zotero to manage their citations and PDFs and want to use LLMs to summarize papers, find related research, or manage their library through natural language.

Highlights

  • Comprehensive Library Access: Search by metadata, full text, or meaning (semantic search) and read PDF page ranges.
  • Deep PDF Integration: Extracts annotations, highlights, and can provide page images when text extraction is poor (e.g., for complex math or tables).
  • Library Management: Ability to add new papers via DOI, URL, ISBN, or BibTeX and manage collections and tags.
  • Flexible Deployment: Supports local-first access for privacy/speed and web API for remote synchronization.
  • Citation Insights: Optional integration with Scite for citation tallies and retraction alerts.

Related

  • Project
    cookjohn/zotero-mcp

    An MCP-based integration that allows AI assistants to search, read, and manage a local Zotero reference library for enhanced academic research.

  • Project
    docsagent/docsagent

    An MCP server that gives AI agents private, local access to search and read Zotero, Obsidian, and Apple Notes using a high-performance C++ search engine.

  • Project
    blazickjp/arxiv-mcp-server

    A local MCP server for agent literature work. The differentiator is original-LaTeX section reads, BibTeX from arXiv metadata, and topic watches. Papers stay on disk. The working loop is paper ID → outline → one section → citations. Search is optional.

  • Project
    neuledge/context

    An MCP server that provides AI agents with instant, local, and up-to-date documentation for 100+ popular libraries to prevent hallucinations caused by outdated training data.

  • Project
    wonderwhy-er/DesktopCommanderMCP

    An MCP server that enables AI assistants to search, edit files, and execute terminal commands on your local machine, with support for advanced file types and process management.