<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> Zotero AI Butler addresses the challenge of overwhelming academic literature by automating the deep reading and summarization of research papers. Targeting researchers and students using Zotero, this plugin leverages large language models to generate concise Markdown notes, enabling users to grasp a paper's essence in minutes and significantly reducing reading time and information overload.

How It Works

The plugin integrates seamlessly with Zotero, offering automated scanning of new literature or batch processing of existing papers. Its core innovation lies in directly processing PDF content using multi-modal LLMs (like Gemini), preserving complex elements such as images, tables, and formulas without relying on OCR. Summaries are generated as structured Markdown notes, accessible within Zotero's interface, and can be enhanced by an experimental "one-image summary" feature that creates an academic poster.

Quick Start & Requirements

• Installation: Download the .xpi file from the project's GitHub Releases page and install via Zotero's "Tools" -> "Add-ons" menu.
• Prerequisites: Zotero desktop client.
• Dependencies: API keys for supported LLM platforms (OpenAI, Gemini, Anthropic, or OpenAI-compatible). Google Gemini 3 Pro is recommended for its PDF understanding capabilities. Users can refer to provided guides for setting up free Gemini API access.
• Setup: Configure API keys and test connectivity within the plugin's settings. Enabling "auto-scan new literature" automates the workflow.

Highlighted Details

• Lossless PDF Processing: Utilizes LLMs' multi-modal capabilities to interpret PDFs directly, including images, tables, and formulas.
• Multi-Platform LLM Support: Integrates with Google Gemini, OpenAI, Anthropic, and OpenAI-compatible services.
• Automated & Batch Processing: Features automatic scanning for new papers and batch processing for backlogs.
• "One-Image Summary": An experimental feature to generate a single-image academic poster summary of a paper.
• Multi-Round Dialogue: Allows for in-depth, conversational analysis and follow-up questions on papers.

Maintenance & Community

The project acknowledges contributors and thanks other open-source projects, notably zotero-ainote, for inspiration and code reference. Specific community channels (e.g., Discord, Slack) or a roadmap are not detailed in the provided README.

Licensing & Compatibility

The README does not specify a software license. This omission requires clarification for potential adoption, especially concerning commercial use or derivative works.

Limitations & Caveats

The "one-image summary" feature is experimental and can incur significant API costs, with automatic generation disabled by default. The effectiveness of summaries is contingent on the chosen LLM and prompt configurations. The absence of a stated license is a critical adoption blocker.