PromptPilot

Better prompts, Better results. Transform your AI interactions with smarter, more effective prompts - Better input, better output.

A Chrome extension that enhances your prompts for ChatGPT, Claude, and Gemini. Choose between 100% local AI processing (no API keys needed) or Gemini API (faster, more powerful).

✨ Features

Dual Processing Modes

• Local AI (WebLLM) - 100% private, no API keys, works offline
• Gemini API - Faster responses, more powerful enhancements (optional)

Core Capabilities

• Universal Platform Support - Works seamlessly on ChatGPT, Claude, and Gemini
• One-Click Enhancement - Transform prompts instantly with structured output
• Privacy First - Local processing option keeps your prompts on your device
• Beautiful UX - Smooth animations, real-time feedback, intuitive interface
• Free Forever - No subscriptions, no usage limits, no catch

Privacy & Security

• Zero Data Collection - We don't collect, store, or track anything
• Local Storage Only - All data stored in your browser
• Open Source - Full transparency, review the code yourself
• Secure API Key Storage - Encrypted local storage (if using Gemini API)

🚀 Quick Start

For Users

1. Install from Chrome Web Store

   • Visit the Chrome Web Store listing
   • Click "Add to Chrome"
   • Grant necessary permissions

2. Choose Your Mode

   • Local AI: No setup needed, just start using!
   • Gemini API: Add your API key in the extension popup (optional)

3. Start Enhancing

   • Navigate to ChatGPT, Claude, or Gemini
   • Type your prompt
   • Click "✨ Enhance"
   • Review and use the enhanced prompt

For Developers

1. Clone the Repository

   git clone <repository-url>
   cd add-in

2. Install Dependencies

   npm install

3. Build the Extension

   npm run build

4. Load in Chrome

   • Open chrome://extensions/
   • Enable "Developer mode"
   • Click "Load unpacked"
   • Select the dist/ directory

5. Development Mode

   npm run dev  # Watch mode for development

📖 Usage

Basic Usage

1. Navigate to any supported AI platform:

   • ChatGPT
   • Claude
   • Google Gemini

2. Type your prompt in the text area

3. Click the "✨ Enhance" button that appears

4. Review the enhanced, structured prompt:

   • Task: Clear objective
   • Context: Relevant background
   • Instructions: Detailed guidelines
   • Output Format: Desired structure

5. Choose an action:

   • Replace: Replace original prompt with enhanced version
   • Copy: Copy enhanced prompt to clipboard
   • Close: Dismiss the enhancement UI

API Key Setup (Optional - Gemini API Mode)

1. Get a free Gemini API key from Google AI Studio
2. Open the extension popup
3. Enter your API key
4. Click "Save Key"
5. Enjoy faster, more powerful enhancements!

🛠️ Development

Project Structure

add-in/
├── src/
│   ├── lib/                  # Shared modules
│   │   ├── constants.js      # Configuration constants
│   │   ├── platform.js       # Platform detection logic
│   │   ├── webllm.js         # WebLLM engine management
│   │   ├── gemini.js         # Gemini API client
│   │   ├── ui.js             # UI components and utilities
│   │   └── utils.js           # General utility functions
│   ├── contentScript.js      # Main content script (runs on AI platforms)
│   ├── service_worker.js     # Background service worker
│   ├── popup.js              # Extension popup script
│   ├── popup.html            # Extension popup UI
│   ├── options.js            # Options page script
│   ├── options.html          # Options page UI
│   ├── summary.html          # Summary page (fallback)
│   ├── privacy.html          # Privacy policy page
│   └── icon*.png             # Extension icons (16, 48, 128)
├── public/
│   └── logo1.png             # Source logo
├── .github/                   # GitHub templates
│   ├── ISSUE_TEMPLATE/
│   └── pull_request_template.md
├── manifest.json              # Extension manifest (Chrome Web Store)
├── package.json               # Dependencies and scripts
├── build.js                   # Build script (esbuild)
├── LICENSE                    # MIT License
└── README.md                  # This file

Build Process

The extension uses esbuild to bundle code and create a production-ready build:

npm run build      # Production build (minified)
npm run dev        # Development build (watch mode)

Build Output:

• Bundled JavaScript files in dist/src/
• Copied HTML, manifest, and assets
• Optimized for production

Architecture

Content Script (contentScript.js)

• Detects AI platform pages (ChatGPT, Claude, Gemini)
• Injects "Enhance" button near text input fields
• Handles prompt enhancement UI
• Manages WebLLM and Gemini API integration
• Processes user interactions (click, copy, replace)

Service Worker (service_worker.js)

• Manages WebLLM engine lifecycle
• Handles model download and caching
• Processes extension messages
• Manages storage operations

Popup (popup.js / popup.html)

• API key management (Gemini API mode)
• Extension information and capabilities
• Simple, focused interface

Options Page (options.js / options.html)

• AI method selection (Local AI vs Gemini API)
• Settings management
• Privacy information

Key Modules

lib/constants.js

• Model configuration
• UI dimensions and styles
• Platform selectors
• Storage keys
• System prompts

lib/platform.js

• Platform detection (ChatGPT, Claude, Gemini)
• Text extraction from platform-specific elements
• Textarea finding logic

lib/webllm.js

• WebLLM engine initialization
• Model download and caching
• Progress tracking
• Engine state management

lib/gemini.js

• Gemini API client
• SSE stream parsing
• Error handling
• API request management

lib/ui.js

• UI element creation
• Enhancement modal management
• Styling utilities
• Button state management

lib/utils.js

• Error message formatting
• Extension context validation
• Safe Chrome API wrappers
• HTML escaping

🔧 Configuration

Model Configuration

Local AI (WebLLM):

• Model: gemma-2-2b-it-q4f16_1-MLC-1k
• Location: src/lib/constants.js - MODEL_ID
• First download: ~2-3GB (one-time)
• Cached in IndexedDB

Gemini API:

• Model: gemini-2.5-flash-lite
• Location: src/lib/gemini.js - GEMINI_MODEL
• Requires API key from Google AI Studio

Permissions

The extension requires:

• storage - Store user preferences and API key locally
• host_permissions - Access to AI platform domains and Gemini API

Note: The extension uses content scripts (declared in manifest.json) which have automatic DOM access. No activeTab, tabs, or scripting permissions are needed.

See CHROME_STORE_PRIVACY_JUSTIFICATIONS.md for detailed permission justifications.

🧪 Testing

Manual Testing

1. Test on Each Platform:

   • ChatGPT: https://chat.openai.com
   • Claude: https://claude.ai
   • Gemini: https://gemini.google.com

2. Test Both Modes:

   • Local AI (WebLLM) - No API key
   • Gemini API - With API key

3. Test Features:

   • Prompt enhancement
   • Copy functionality
   • Replace functionality
   • Error handling
   • UI responsiveness

Browser Requirements

• Chrome 113+ or Edge 113+ (WebGPU support required)
• For Chrome 123, enable: chrome://flags/#enable-experimental-web-platform-features
• 4GB+ RAM recommended
• Stable internet connection for initial model download (Local AI mode)

🐛 Troubleshooting

WebGPU Not Supported

• Update to Chrome 113+ or Edge 113+
• Enable chrome://flags/#enable-experimental-web-platform-features
• Restart browser

Model Loading Issues (Local AI)

• Ensure stable internet connection for first download
• Check browser console for errors
• Clear extension storage and retry
• Verify WebGPU is enabled

Gemini API Issues

• Verify API key is correct
• Check API key has proper permissions
• Ensure internet connection is active
• Check browser console for error messages

Extension Not Working

• Reload the extension (chrome://extensions)
• Refresh the AI platform page
• Check browser console for errors
• Verify you're on a supported platform

📝 Contributing

We welcome contributions! Please see our contributing guidelines:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Make your changes
4. Test thoroughly
5. Commit your changes (git commit -m 'Add amazing feature')
6. Push to the branch (git push origin feature/amazing-feature)
7. Open a Pull Request

Development Guidelines

• Follow existing code style
• Add comments for complex logic
• Test on all supported platforms
• Update documentation as needed
• Ensure error handling is robust

🗺️ Roadmap

Coming Soon (v2.0)

• Personalized Prompt Personas
  • Example-Oriented Persona
  • Essay Persona
  • Depth-Focused Persona
  • Concise Persona
  • Creative Persona
  • Technical Persona
  • Business Persona

Future Features

• Prompt templates library
• Prompt history and management
• Custom enhancement rules
• Keyboard shortcuts
• Batch processing
• Team collaboration features

📄 License

MIT License - see LICENSE file for details.

🙏 Acknowledgments

• WebLLM by MLC AI for local LLM inference
• Google Gemini API for API-based enhancement
• Chrome Extensions team for the excellent platform

📞 Support

• Issues: GitHub Issues
• Documentation: See DEVELOPER.md for detailed developer docs
• Privacy: See src/privacy.html for privacy policy

---

Made with ❤️ for the AI community