An intelligent writing assistant that uses AI to continue your text seamlessly. Built with React, XState, ProseMirror, and Radix UI.

• ✍️ AI-Powered Writing: Continue your text with AI-generated content
• ⌨️ Keyboard Shortcuts: Quick access with Ctrl+Space (or Cmd+Space on Mac)
• 🎨 Rich Text Editor: Powered by ProseMirror with formatting support
• 🔄 State Management: Robust state handling with XState
• 🎯 Content Control: Accept, clear, or discard AI suggestions
• 📱 Responsive Design: Works on desktop and mobile
• ♿ Accessible: Built with Radix UI for WCAG compliance

• React 18: Modern UI library with hooks
• XState v5: State machine for predictable state management
• ProseMirror: Powerful rich text editor framework
• Radix UI: Accessible, unstyled UI primitives
• OpenAI API: GPT-4 for content generation
• Vite: Fast build tool and dev server

• Node.js 16+ and npm
• OpenAI API key (Get one here)

1. Clone the repository

   git clone <repository-url>
   cd chronicle_assignment_sheetal_singh

2. Install dependencies

   npm install

3. Configure API Key

   Create a .env file in the project root:

   cp .env.example .env

   Edit .env and add your OpenAI API key:

   VITE_OPENAI_API_KEY=your_openai_api_key_here
   

4. Start the development server

   npm run dev

5. Open in browser

   Navigate to http://localhost:5173

1. Start typing in the editor
2. Click "Continue Writing" or press Ctrl+Space (Windows/Linux) or Cmd+Space (Mac)
3. Wait for AI to generate content
4. Review the highlighted AI-generated text
5. Choose an action:
   • Accept All: Keep the AI content
   • Clear All: Remove the AI content
   • Discard Session: Revert to before AI generation

• Ctrl+Space / Cmd+Space: Trigger AI generation
• Ctrl+Z / Cmd+Z: Undo
• Ctrl+Y / Cmd+Y: Redo

src/
├── components/
│   ├── Editor.jsx        # ProseMirror editor component
│   ├── Editor.css        # Editor styles
│   ├── Toolbar.jsx       # Main toolbar with Continue button
│   ├── StatusBar.jsx     # Status indicator at bottom
│   └── AIToolbar.jsx     # Floating toolbar for AI actions
├── machines/
│   └── editorMachine.js  # XState state machine
├── services/
│   └── aiService.js      # OpenAI API integration
├── App.jsx               # Main application component
└── App.css               # Global styles

The application uses XState to manage states:

idle → generating → review → idle
  ↓                    ↓
error ←──────────────┘

• idle: Ready for user input
• generating: AI is creating content
• review: AI content ready for user action
• error: Something went wrong

All code files include extensive comments explaining:

• What each function does
• Why certain approaches were chosen
• How different parts interact
• React and JavaScript concepts for beginners

npm run build

npm run preview

npm run lint

⚠️ Important: This demo stores the API key in the frontend for simplicity. In production:

1. Never expose API keys in frontend code
2. Use a backend server to make API calls
3. Implement proper authentication
4. Add rate limiting

• Local AI fallback with transformers.js
• Multiple AI model selection
• Customizable AI parameters (temperature, max tokens)
• Export to Markdown/PDF
• Dark mode support
• Collaborative editing
• Version history with branching

• React Documentation
• XState Documentation
• ProseMirror Guide
• Radix UI
• OpenAI API

MIT

Sheetal Singh