Holo is a full-featured cross-platform private Retrieval-Augmented Generation (RAG) system that provides intelligent knowledge management, document processing, and AI Agent conversation capabilities.

✨ Features

🌐 Cross-Platform Support

• Web Application: Modern web interface built with Next.js
• Desktop Application: Native desktop experience powered by Tauri
• Mobile Friendly: Responsive design supporting mobile device access
• Consistent Experience: Unified user interface and functionality across platforms

🧠 Intelligent Knowledge Management

• Multi-modal Information Extraction: Support for documents, images, audio, and video information extraction
• Smart Chunking: Automatically split documents into optimal knowledge blocks
• Vector Retrieval: Efficient semantic search and similarity matching
• Knowledge Graph: Build structured knowledge relationship networks

📄 Document Processing

• Support for multiple document formats: PDF, Word, PowerPoint, CSV, HTML, Markdown
• Automatic content recognition and structured processing
• Intelligent Q&A generation and preview functionality
• Document chunking and vectorized storage

💬 AI Conversation System

• Retrieval-augmented intelligent conversations
• Support for multiple AI model integrations
• Real-time Markdown rendering
• Code highlighting and syntax support

🎨 Creative Tools

• Integrated Novel editor with rich text editing support
• AI-assisted writing and content generation
• Mathematical formulas and chart support
• Real-time collaboration and version control

🔧 Multi-Agent System

• Planner Agent: Task planning and decomposition
• Coder Agent: Code generation and analysis
• Reporter Agent: Report generation and BI analysis
• Researcher Agent: Research and information gathering

🏗️ Technical Architecture

Frontend Stack

• Framework: Next.js 14 + TypeScript
• UI Components: Tailwind CSS + shadcn/ui
• Editor: TipTap + Novel
• State Management: Zustand
• Form Handling: React Hook Form + Zod

Backend Stack

• Framework: FastAPI + Python
• AI Integration: LlamaIndex + Ollama
• Vector Database: ChromaDB
• File Storage: Local storage + multi-format support
• Memory Management: File system caching

Desktop Application

• Cross-Platform: Tauri + Rust
• Local Deployment: Completely offline operation
• System Integration: Native desktop experience

🚀 Quick Start

Prerequisites

• Node.js 18+
• Python 3.12+
• Rust (for Tauri)

Install Dependencies

# Frontend dependencies
npm install

# Backend dependencies
cd backend
pip install -r requirements.txt

Start Development Environment

# Start frontend development server
npm run dev

# Start backend service
cd backend
uvicorn main:app --reload

# Start Tauri desktop application
npm run tauri dev

Build for Production

# Build frontend
npm run build

# Build desktop application
npm run tauri build

📁 Project Structure

holo/
├── src/                          # Frontend source code
│   ├── app/                      # Next.js page routing
│   ├── components/               # Common components
│   ├── features/                 # Feature modules
│   │   ├── extraction/           # Information extraction
│   │   ├── knowledge/            # Knowledge management
│   │   ├── editor/               # Editor
│   │   └── chunk/                # Chunking processing
│   └── services/                 # API services
├── backend/                      # Backend source code
│   ├── agent/                    # Agent system
│   ├── prompt/                   # Prompt templates
│   └── services/                 # Business services
├── src-tauri/                    # Tauri desktop application
└── public/                       # Static assets

🔧 Main Features

Knowledge Base Management

• Create and manage multiple knowledge bases
• Document upload and automatic processing
• Intelligent chunking and vectorization
• Retrieval testing and optimization

Multi-modal Processing

• Document information extraction: DocInformationExtraction
• Image information extraction: PictureInformationExtraction
• Multi-media processing: MultiModalInformationExtraction

AI Conversation

• Intelligent Q&A system
• Contextual understanding
• Multi-turn conversation support
• Markdown rendering: MarkdownText

Code Editing

• Syntax highlighting
• Code completion
• Real-time preview
• Multi-language support

📱 Platform Support

• 🖥️ Desktop: Windows 10+, macOS 10.15+, Linux (Ubuntu 18.04+)
• 🌐 Web: Modern browsers (Chrome 88+, Firefox 85+, Safari 14+)
• 📱 Mobile: iOS and Android support via PWA