Zotero

The Zotero MCP Connector is a tool for integrating the local Zotero literature library with the Claude desktop version, enabling direct reading and access through the local Web API.

Research and data Knowledge management and memory #Literature management #API integration #Scientific research tools .Python

rating : 2.5 points

downloads : 6.8K

update time : 2025-04-28

Open Site

What is the Zotero MCP Connector?

The Zotero MCP Connector is a tool that connects the local Zotero library with the Claude desktop application. It is implemented through the Model Context Protocol (MCP), allowing users to directly search for and access their literature materials.

How to use the Zotero MCP Connector?

After installation, you can configure and start using it through simple commands. You can synchronize your local Zotero library with Claude in just a few steps.

Applicable scenarios

Suitable for researchers, students, or professionals who need to efficiently manage their personal literature resources.

Main features

Get all collections

List all collections in the local Zotero library.

Get the contents of a specific collection

View the details of the literature in a specific collection.

Get literature details

View the specific information of a piece of literature, such as the abstract.

Search the entire library

Conduct a keyword search in your entire Zotero library.

Get recently added literature

Quickly get the literature added in the past 10 days.

Advantages

Seamlessly integrate Zotero with Claude to improve work efficiency.

No need to manually import literature data, with real-time synchronization.

Support multiple literature retrieval methods to meet diverse needs.

Limitations

Only support personal libraries, and team sharing is not supported yet.

Ensure that the Zotero version is 7 or higher.

Dependent on the network environment, which may affect the response speed.

How to use

Install the UV tool

Download and install the UV tool for running the MCP server.

Clone the project code

Clone the project code from GitHub to your local machine.

Enable the Zotero local API

Enable the local API in the Zotero settings.

Install the MCP plugin

Install the Zotero MCP plugin through the UV command.

Usage examples

Example 1: Get all collections

Show how to list all collections in the local Zotero library.

Example 2: Search for literature

Demonstrate how to search for literature with a specific keyword.

Frequently Asked Questions

Does it support team-shared libraries?

How to solve the installation failure problem?

Can I customize the search scope?

Related resources

GitHub repository

The source code and documentation of the project.

Claude MCP documentation

Understand the basic knowledge of the MCP protocol.

🚀 Zotero MCP Connector

This is a Model Control Protocol (MCP) connector that integrates your local Zotero with Claude. It allows you to directly access your local Zotero library through Claude's desktop interface. This connector relies on the availability of the local Web API in Zotero 7.

This project is inspired by the mcp-zotero repository using Node.js and the Web API. It is built on the excellent pyzotero library.

🚀 Quick Start

This is a Model Control Protocol (MCP) connector designed to integrate your local Zotero with Claude. It enables you to directly access your local Zotero library via Claude's desktop interface. This connector relies on the local Web API available in Zotero 7.

📦 Installation

For information on how Claude Desktop interacts with MCP, refer to here.

Use uv. For installation instructions, visit here.
Clone the project locally and activate the virtual environment:

git clone https://github.com/gyger/mcp-pyzotero.git
cd mcp-pyzotero
uv sync

Enable the local API in Zotero 7:
Add the server to your local Claude installation:

uv run mcp install zotero.py

📚 Documentation

Configuration

The connector is configured to work with a local Zotero installation and currently only supports user libraries. By default, it uses the user ID 0, but you can also specify it by setting the environment variable ZOTERO_USER_ID:

uv run mcp install zotero.py -v ZOTERO_USER_ID=0

Available Features

The connector provides the following features:

get_collections(): List all collections in your Zotero library.
get_collection_items(collection_key): Get all items in a specific collection.
get_item_details(item_key): Get detailed information about a specific paper, including the abstract.
search_library(query): Search the entire Zotero library.
get_recent(limit=10): Get the papers recently added to the library.

Future plans include expanding these features.