Your photographic browsing memory. Search everything you've seen online.
BrowseBack is a Chrome extension that automatically captures and indexes your browsing activity, creating a searchable "photographic memory" of everything you've seen online. All data processing happens locally using Chrome's built-in AI, ensuring complete privacy.
|
|
|
|
|
|
|
- 🎯 The Problem
- ✨ The Solution
- 🔒 Privacy First
- 👥 Use Cases
- 🚀 Installation
- 🎮 Usage
- 🏗️ Architecture
- 🎓 How It Works
- 📚 Documentation
- 🛠️ Development
- 🗺️ Roadmap
- 🏆 Google Chrome Built-in AI Challenge 2025
- 🎬 Demo Video
- ⚙️ Configuration
- 🤝 Contributing
- 📄 License
- 💡 FAQ
- Users forget where they saw information online
- Browser history is just a list of URLs (not searchable by content)
- Can't remember context: "What was I researching Tuesday afternoon?"
- Screenshots/bookmarks require manual effort (and people forget)
BrowseBack runs silently in the background, automatically:
- 📸 Capturing screenshots of your active tab
- 🔤 Extracting text using OCR powered by Chrome Prompt API
- 🔍 Building a searchable local index of all content
- 🧠 Letting you search by content, time, or natural language
- ✅ 100% local processing - your browsing data never leaves your device
- ✅ Works completely offline - no internet required for search
- ✅ Zero subscription fees - no cloud costs
- ✅ You own your data - export or delete anytime
BrowseBack is perfect for anyone who needs to recall what they've seen online:
- 🎓 Students - Find lecture notes, research papers, and study materials you've browsed
- 💼 Executives - Recall details from market research, competitor analysis, and reports
- 👨💻 Engineers - Quickly find API docs, Stack Overflow solutions, and technical articles
- 💰 Sales Teams - Remember product details, customer research, and competitive intel
- 🧠 People with ADHD - Never lose track of important information in your browsing sessions
- 🔬 Researchers - Track down sources and references you encountered during your work
- 📝 Content Creators - Find inspiration and reference materials from your browsing history
- Google Chrome (version 127+)
- Chrome Built-in AI Early Preview Program enrollment
- Sign up at: Chrome Built-in AI
- Enable flags:
chrome://flags/#optimization-guide-on-device-modelchrome://flags/#prompt-api-for-gemini-nano
-
Clone this repository
git clone https://github.com/yourusername/browseback.git cd browseback -
Load in Chrome
- Open Chrome and navigate to
chrome://extensions/ - Enable Developer mode (toggle in top right)
- Click Load unpacked
- Select the
BrowseBackfolder
- Open Chrome and navigate to
-
Grant Permissions
- Click on the BrowseBack icon in your toolbar
- Allow necessary permissions when prompted
-
Download AI Models (First Time Setup)
- When you first open BrowseBack, you'll see a setup guide
- Click "Download AI Components" to start downloading Gemini Nano
- The download is ~1.5GB and may take several minutes
- You can track progress in the extension popup
- Note: Models download once and are shared across all Chrome AI apps
-
Start Browsing!
- Once AI models are ready, BrowseBack starts capturing automatically
- Click the extension icon anytime to search your memory
- Use Timeline, Search, or Ask modes to find what you need
💡 Tip: For detailed AI setup instructions, see the AI Setup Guide
- BrowseBack captures your active tab every 10 seconds (configurable)
- Only captures when tab content changes (saves storage)
- Skips internal Chrome pages (
chrome://,edge://, etc.)
- Click the BrowseBack icon in your toolbar
- Type your search query:
- Content search: "Python tutorial" finds pages with that text
- Time-based: Click "Today" or "Yesterday" chips
- URL search: Search by domain or path
- Blank search: Shows recent captures
- View total captures and storage used
- Adjust capture interval (5s to 1 minute)
- Set data retention period (3 days to 1 year)
- Delete all data with one click
BrowseBack/
├── manifest.json # Chrome extension configuration
├── background.js # Service worker (automatic capture)
├── content.js # Content script (DOM text extraction)
├── lib/
│ ├── storage.js # IndexedDB wrapper
│ ├── capture.js # Screenshot capture logic
│ └── ai-processor.js # Chrome Prompt API integration
└── popup/
├── popup.html # Search interface
├── popup.js # UI logic
└── popup.css # Styling
- Chrome Extension Manifest V3
- Google Gemini Nano - On-device AI model powering all intelligent features
- Chrome Prompt API (LanguageModel) - AI question answering, transcription, and summarization
- Chrome Proofreader API - Query correction and spell checking
- Chrome Speech Recognition API - Real-time audio capture from tabs
- IndexedDB - Local storage for captures and transcripts
- WebP compression - Efficient screenshot storage
- Service Workers - Background processing and capture automation
- Capture: Background service worker monitors active tab
- Change Detection: Only captures when content changes significantly
- Screenshot: Uses
chrome.tabs.captureVisibleTab()API - Text Extraction:
- Primary: DOM text extraction (fast, accurate)
- Secondary: OCR via Prompt API (images, videos)
- Storage: Saves to IndexedDB with metadata (URL, title, timestamp)
- Search: Full-text search across all stored content
- Cleanup: Automatically deletes old data based on retention settings
BrowseBack solves the universal problem of digital forgetfulness. Users spend hours re-searching for information they've already seen. Our solution:
- Unlocks photographic memory for everyone
- Privacy-first approach (no cloud, no tracking)
- Zero friction (automatic, passive capture)
- Universal use case (students, researchers, developers, anyone)
Unlike cloud-based solutions, BrowseBack is:
- ✅ 100% free (no server costs)
- ✅ 100% private (local-only processing)
- ✅ 100% offline (network-resilient)
Additional documentation is available in the docs/ directory:
- Quickstart Guide - Get started quickly
- Architecture - Technical deep dive
- AI Setup Guide - Chrome Built-in AI setup
- Demo Script - 3-minute demo walkthrough
- Testing Guide - How to test the extension
- Submission Guide - Hackathon submission details
# Clone the repo
git clone https://github.com/yourusername/browseback.git
cd browseback
# No build step required! Pure vanilla JS.- Make your changes
- Go to
chrome://extensions/ - Click "Reload" on BrowseBack
- Test in a new tab
- Service Worker:
chrome://extensions/→ BrowseBack → "service worker" link - Popup: Right-click extension icon → "Inspect popup"
- Content Script: Regular DevTools on any page
This is a hackathon project, but contributions are welcome!
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE file for details.
- Google Chrome Team for the Built-in AI Challenge 2025
- Chrome Prompt API for enabling on-device AI
- Devpost for hosting the hackathon
- All early testers and contributors
Q: Does BrowseBack slow down my browser? A: No. Captures happen asynchronously and don't block tab rendering.
Q: Can I use this on mobile? A: Not yet. Chrome extensions don't support mobile, but we're exploring Firebase AI.
Q: What happens if I run out of storage? A: BrowseBack automatically deletes old captures based on your retention settings.
Q: Can I export my data? A: Yes! Use the browser's IndexedDB export tools, or we'll add a built-in export feature soon.
Q: Is this really private? A: Absolutely. All data stays in IndexedDB on your device. No network requests, no telemetry.
This project was created for the Google Chrome Built-in AI Challenge 2025 to showcase the power of client-side AI and privacy-first innovation.
Made with 🧠 and ❤️ by Samuel Donkor
Give your browser a brain boost!