AI Writer - Local Writing Assistant

A modern desktop application combining rich text editing with AI-powered content generation and transformation.

Demo

Key Features

• 🤖 Multi-Provider AI Support - OpenAI, Anthropic, or Local Qwen models
• ✍️ Rich Text Editor - Professional editing powered by Lexical
• ⚡ Real-Time AI Streaming - Live content generation and transformation
• 🎯 Transformation Presets - Save and reuse custom text transformation prompts
• 💾 Persistent Workspace - Automatic saving of editor content and settings
• 🌓 Dark/Light Themes - Comfortable writing in any lighting condition
• 📊 System Monitoring - AI model status, application logs, and system information
• 🌍 Internationalization - Multi-language support (English, French)
• 🔒 Privacy-First - Desktop app with local storage and optional local AI models

Capabilities

• Generate content from prompts with streaming AI responses
• Transform selected text with AI (e.g., "Make it professional", "Simplify", "Expand")
• Configure and switch between AI providers with adjustable parameters
• Monitor AI model status, logs, and system specs

Quick Start

Prerequisites

• Node.js v18 or higher
• pnpm v8 or higher
• Rust latest stable
• Tauri Prerequisites for your platform

Installation

# Clone the repository
git clone https://github.com/marouane-dev75/ai-writer
cd ai-writer

# Install dependencies
pnpm install

# Run in development mode
pnpm tauri dev

Build for Production

# Build the application
pnpm tauri build

The built application will be available in src-tauri/target/release/.

Technology Stack

Frontend

• React 19 - Modern UI library with latest features
• TypeScript - Type-safe development
• Tailwind CSS 4 - Utility-first styling
• Lexical - Extensible rich text editor framework
• Vite - Fast build tooling
• React Router 7 - Client-side routing

Backend

• Rust - High-performance systems language
• Tauri 2 - Secure desktop application framework
• Serde - Serialization/deserialization
• Tokio - Async runtime for AI streaming

AI Integration

• OpenAI API - GPT models
• Anthropic API - Claude models
• Local Qwen - Privacy-focused local models

Project Structure

ai-writer/
├── src/                          # Frontend application
│   ├── features/                 # Feature modules
│   │   ├── ai-runtime/          # AI streaming & status monitoring
│   │   ├── ai-settings/         # AI provider configuration
│   │   ├── editor/              # Rich text editor with AI
│   │   ├── logging/             # Application logging viewer
│   │   ├── system-info/         # System information display
│   │   └── app-restart/         # Application restart management
│   ├── pages/                   # Page components
│   ├── shared/                  # Shared utilities & components
│   │   ├── ui/                  # Reusable UI components
│   │   ├── i18n/                # Internationalization
│   │   ├── theme/               # Theme management
│   │   └── layouts/             # Layout components
│   └── main.tsx                 # Application entry point
│
└── src-tauri/                   # Backend application
    └── src/
        ├── ai/                  # AI provider implementations
        ├── commands/            # Tauri command handlers
        ├── config/              # Configuration management
        ├── editor/              # Editor persistence
        └── logging/             # Logging system

Feature Documentation

Each feature has detailed documentation in its respective directory:

• AI Runtime - AI streaming and model status
• AI Settings - Provider configuration
• Editor - Rich text editing with AI
• Logging - Application log viewer
• System Info - System specifications
• App Restart - Restart management

Development

Available Commands

# Frontend development server
pnpm dev

# Tauri development (with hot-reload)
pnpm tauri dev

# Build frontend
pnpm build

# Build Tauri application
pnpm tauri build

# Preview production build
pnpm preview

Code Organization

This project follows clean architecture principles:

• Feature-based structure - Each feature is self-contained
• Dependency Inversion - High-level modules depend on abstractions
• SOLID principles - Maintainable and extensible code
• TypeScript strict mode - Type safety throughout
• Rust best practices - Idiomatic Rust with proper error handling

See individual feature READMEs for detailed implementation guidelines.