This project offers an AI-driven tool for parsing EPUB and PDF ebooks, transforming them into structured mindmaps and concise text summaries. It targets users seeking efficient knowledge extraction and organization from digital books, providing a streamlined method to digest complex content.

How It Works

The tool leverages Google Gemini or OpenAI GPT models for content processing. It supports three distinct modes: generating chapter-specific mindmaps, creating a comprehensive mindmap for the entire book, or producing detailed text summaries. Key features include intelligent chapter detection, the ability to skip non-essential sections (like prefaces or indexes), and an efficient caching mechanism to avoid redundant AI computations, optimizing processing time.

Quick Start & Requirements

• Installation: Clone the repository, then run pnpm install (or npm install).
• Prerequisites: Node.js 18+ and pnpm (recommended) or npm.
• Execution: Start the development server with pnpm dev (or npm run dev). Access the application at http://localhost:5173.
• Configuration: Requires API keys for Google Gemini or OpenAI, configured via a settings panel.

Highlighted Details

• Supports EPUB and PDF file formats.
• Integrates with Google Gemini and OpenAI GPT.
• Offers flexible output: chapter summaries, chapter mindmaps, or full-book mindmaps.
• Features intelligent chapter parsing, including automatic detection and skipping of irrelevant content.
• Includes an efficient local caching system for processed results.
• Provides export functionality for mindmaps (PNG, SVG) and summaries (Markdown, TXT).

Maintenance & Community

The README outlines contribution guidelines, welcoming Issues and Pull Requests, and specifies development environment setup and code linting. No specific details regarding maintainers, community channels, or project roadmaps are provided.

Licensing & Compatibility

The project is licensed under the permissive MIT License, allowing for broad usage and modification, including commercial applications. No specific compatibility restrictions are noted beyond browser recommendations.

Limitations & Caveats

PDF parsing may be unreliable for encrypted or image-based documents. A recommended file size limit of 50MB is stated. Caching relies on browser localStorage, which has inherent storage limitations. Stable network connectivity is required for AI API interactions.