🦖 enzonic Docasaurus

enzonic Docasaurus is a powerful, green-themed web application designed to help you process and modify large PDF documents using AI. It breaks down complex documents into manageable chunks, processes them via AI with custom prompts, and allows you to track progress in real-time.

✨ Features

• 📑 PDF Intelligent Chunking: Automatically splits large PDFs into smaller text blocks for efficient AI processing.
• 🤖 AI-Driven Modifications: Deep integration with OpenAI-compatible APIs (like Chutes.ai) to transform your text based on custom prompts.
• 🔄 Real-time Progress Tracking: Watch your document being processed piece-by-piece with a live dashboard.
• 🛡️ Resilience & Autoretry: Includes a built-in 10-second retry mechanism for any failed AI requests.
• 💾 Persistence: Tasks are saved to a local database, allowing you to resume interrupted work even after a server restart.
• 🍏 Modern Green UI: A sleek, user-friendly interface built with React and Vite.

🚀 Getting Started

Prerequisites

• Node.js (v18 or higher)
• An AI API Key (OpenAI or compatible provider like Chutes.ai)

Installation

1. Clone the repository:

   git clone https://github.com/enzonic-llc/docasaurus
   cd docasaurus

2. Install dependencies:

   npm install

   (This will install dependencies for both the root, client, and server)

3. Configure Environment Variables: Copy the .env.example file to .env and fill in your details:

   cp .env.example .env

   Edit .env:

   • OPENAI_API_KEY: Your secret key.
   • OPENAI_COMPLETIONS_URL: Set to any OpenAI compatible endpoint or the standard OpenAI URL.
   • MODEL_ID: e.g., gpt4o or any model the api supports

Running the App

To start both the frontend and backend in development mode:

npm run dev

• Frontend: http://localhost:5173
• Backend: http://localhost:5000

🛠️ Tech Stack

• Frontend: React, Vite, Axios, Lucide Icons.
• Backend: Node.js, Express, Multer (File Uploads), pdf-parse.
• Database: lowdb (JSON-based local persistence).
• Styling: Vanilla CSS with a focus on modern aesthetics.

📜 License

Distributed under the MIT License. See LICENSE for more information.

🙌 Acknowledgments

• Built with ❤️ by the enzonic team.
• Powered by the latest in agentic AI.