Zotero Automation

A Python tool for analyzing Zotero library tags using AI categorization and visualization. Use at your own risk.

Features

• Extract and analyze tags from Zotero SQLite database
• AI-powered tag categorization via cborg, OpenAI, or Anthropic
• Generate word clouds from tags
• Export categorized tags in Obsidian markdown format
• Save unique tags for further analysis

Installation

1. Clone the repository:

git clone https://github.com/gnzng/zotero_automate.git
cd zotero_automate

2. Install dependencies:

pip install -r requirements.txt

3. Set up environment variables:

cp .env-example .env
# Edit .env with your actual values

You need the following variables in your .env:

• ZOTERO_DB_PATH: Path to your Zotero SQLite database
• BACKEND: One of cborg (default), openai, or anthropic

Depending on the backend, set the corresponding variables:

Backend	Required variables
cborg	CBORG_API_KEY, CBORG_BASE_URL, CBORG_MODEL
openai	OPENAI_API_KEY, OPENAI_MODEL
anthropic	ANTHROPIC_API_KEY, ANTHROPIC_MODEL

cborg (https://cborg.lbl.gov/) is an OpenAI-compatible API and is the default backend.

Usage

Make sure zotero is closed, since we need database access. Run the main analysis:

python main.py

You should be able to uncomment or comment the things you would like to run your analysis on.

Output Files

• unique_tags.txt: List of unique tags from your library
• categorized_tags.md: AI-categorized tags in markdown format, that works great to include in obsidian
• Word cloud visualization

• Interactive network visualizer from tags and categories

• Tag categories as an interactive radar plot

Contributing

Please feel free to open a PR or issue, I am looking forward to feedback.