
mcp pyzotero
A Model Control Protocol (MCP) connector for integrating your local Zotero with Claude
Repository Info
About This Server
A Model Control Protocol (MCP) connector for integrating your local Zotero with Claude
Model Context Protocol (MCP) - This server can be integrated with AI applications to provide additional context and capabilities, enabling enhanced AI interactions and functionality.
Documentation
Zotero MCP Connector
A Model Control Protocol (MCP) connector for integrating your local Zotero with Claude.
This enables direct read access to your local Zotero library through Claude's Desktop interface.
It depends on the ability to access a local web-api in Zotero 7.
This was inspired by a repository using Node.js and the web api: mcp-zotero.
This builds on the shoulders of the fantastic pyzotero library.
Installation
Run from local code (Recommended)
Information about Claude Desktop interacting with MCPs can be found here.
-
Use
uv. Installation instructions can be found here. -
Checkout the git project to local space and activate the virtual environment inside:
git clone https://github.com/gyger/mcp-pyzotero.git
cd mcp-pyzotero
uv sync
-
Enable the local API in Zotero 7: !Zotero Local API Settings
-
Add the server to your local Claude installation:
uv run mcp install zotero.py
Run encapsulated with uvx (Should work)
Edit the configuration for your Claude Desktop softare in the file.
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
and add the Zotero entry
{
"mcpServers": {
"Zotero": {
"command": "uvx",
"args": ["--from", "git+https://github.com/gyger/mcp-pyzotero.git",
"--with", "mcp[cli]",
"--with", "pyzotero",
"mcp", "run", "zotero.py"
],
}
}
}
Configuration
The connector is configured to work with local Zotero installations and currently only user libraries are supported.
By default it uses the userid 0, but you can also set the environment variable ZOTERO_USER_ID if needed:
uv run mcp install zotero.py -v ZOTERO_USER_ID=0
Available Functions
Available tools
get_zotero_summary(): Lists properties about your library including collections, recent items or tags.get_collection_items(collection_key): Get all items in a specific collectionget_items_metadata(item_key): Get detailed information about specific paper(s), including abstract.search_library(query, mode): Search your Zotero library, with two possible modes: everything or titleCreatorYear.
This functionality should be extended in the future.
Requirements
- Python 3.10+
- pyzotero
- mcp[cli]
- Local Zotero installation
Contributing
Contributions are welcome! Please visit the GitHub repository to:
- Report issues
- Submit pull requests
- Suggest improvements
License
MIT
Quick Start
Clone the repository
git clone https://github.com/gyger/mcp-pyzoteroInstall dependencies
cd mcp-pyzotero
npm installFollow the documentation
Check the repository's README.md file for specific installation and usage instructions.
Repository Details
Recommended MCP Servers
Discord MCP
Enable AI assistants to seamlessly interact with Discord servers, channels, and messages.
Knit MCP
Connect AI agents to 200+ SaaS applications and automate workflows.
Apify MCP Server
Deploy and interact with Apify actors for web scraping and data extraction.
BrowserStack MCP
BrowserStack MCP Server for automated testing across multiple browsers.
Zapier MCP
A Zapier server that provides automation capabilities for various apps.