herbertkokholm/cite-caddy

Overview

A standalone remote MCP server that provides full read/write access to a Zotero library. It runs as a separate Python service and exposes Zotero's data through the Model Context Protocol.

Best for

Best for
Developers building AI-driven workflows that need to manage Zotero collections directly.

Use cases

• Add, update, and delete Zotero items programmatically
• Fetch references and metadata from Zotero via an MCP client
• Integrate Zotero library management into AI workflows

How to use

Install

uv pip install -e ".[dev]"

Tools exposed

• search_items
• get_item
• list_collections
• list_tags
• list_trash
• list_saved_searches
• list_groups
• list_item_types
• list_item_fields
• list_item_type_fields
• list_item_creator_types
• list_creator_fields
• list_attachments
• get_fulltext
• download_attachment
• list_notes
• export_bibliography
• create_item
• create_collection
• create_saved_search

Tested with

Claude Desktop

Notes

A standalone remote MCP server that provides full read/write access to a Zotero library. It runs as a separate Python service and exposes Zotero’s data through the Model Context Protocol.

0 stars on GitHub. Last updated 2026-09-07. Licensed MIT.

Use cases

• Add, update, and delete Zotero items programmatically
• Fetch references and metadata from Zotero via an MCP client
• Integrate Zotero library management into AI workflows

Pros

• Offers full read/write access, not just read-only
• Runs as a standalone remote server without needing a Zotero plugin
• Python-based, easy to extend or integrate into existing tooling

Cons

• No community traction or stars yet, so adoption risk is high
• Requires separate server setup and configuration
• Depends on Zotero’s API, so performance is limited by that service

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.