CapSummarize is a free, open-source browser extension that extracts video captions and helps you generate AI summaries, thumbnails, and short video clips using your favorite AI providers (ChatGPT, Claude, Gemini, Grok, and more).
- Extract captions from YouTube, Google Drive, Udemy, Zoom, and more
- Multiple summary styles - Choose from 15+ different formats (Default, Educational, Technical, Podcast, Kids-friendly, etc.)
- π¨ AI Image Generation - Analyze transcript to create thumbnails, infographics, mind maps, and more
- π¬ AI Video Generation - Analyze transcript to generate video clips (Gemini only)
- Use any AI provider - ChatGPT, Claude, Gemini, Grok, Perplexity, Mistral, Meta AI, and more
- Custom variants - Create your own summary templates
- History tracking - Access previously captured transcripts
- Privacy-focused - All processing happens locally, no data sent to external servers
- 100% Free - No subscriptions, no limits
See what CapSummarize can generate! All examples below were created from real YouTube videos.
| Style | Without Reference | With Reference |
|---|---|---|
| MrBeast Style | ![]() |
![]() |
| Casey Neistat Style | ![]() |
![]() |
| Theo/Tech Style | ![]() |
![]() |
| 5-Min Crafts Style | ![]() |
![]() |
| General Thumbnail | ![]() |
![]() |
| Infographic | Mind Map | Comic Strip |
|---|---|---|
![]() |
![]() |
![]() |
| Whiteboard | Quote Card | Scene |
|---|---|---|
![]() |
![]() |
![]() |
| Style | Description | Example |
|---|---|---|
| Default | Balanced, comprehensive summary | View |
| Educational | Academic-focused with key concepts | View |
| Technical | Code-focused with implementation details | View |
| Casual | Friendly, conversational tone | View |
| Executive | Business-focused brief | View |
| Blog | SEO-optimized blog post | View |
| X/Twitter | Thread-ready format | View |
| Cheatsheet | Quick reference card | View |
| Kids | Simple language for ages 7-12 | View |
-
Clone the repository
git clone https://github.com/samuxbuilds/capsummarize.git cd capsummarize/capsummarize-frontend -
Install dependencies (using Bun recommended)
bun install # or npm install -
Build the extension
bun run build # or npm run build -
Load in Chrome
- Open
chrome://extensions/ - Enable Developer mode (top-right)
- Click Load unpacked and select the
dist/folder
- Open
Start the development server with hot reload:
bun run dev
# or
npm run devThis watches for file changes and rebuilds automatically.
| Command | Description |
|---|---|
npm run dev |
Start development mode with watch |
npm run build |
Build for production |
npm run lint |
Run ESLint |
npm run format |
Format code with Prettier |
npm run type-check |
Run TypeScript type checking |
npm run clean |
Remove build artifacts |
src/
βββ background/ # Service worker modules
β βββ messageHandlers.ts
β βββ sidePanelManager.ts
β βββ storageHelpers.ts
β βββ vttCacheManager.ts
βββ config/ # Configuration
β βββ prompts.ts # Summary prompt templates
β βββ providers.ts # AI provider definitions
βββ services/ # Business logic
β βββ providerService.ts
βββ types/ # TypeScript type definitions
βββ ui/ # User interface
β βββ components/ # Reusable UI components
β βββ state/ # State management
β βββ sidepanel.ts # Main side panel
β βββ settings.ts # Settings page
βββ utils/ # Utility functions
β βββ storage.ts # IndexedDB wrapper
β βββ vtt.ts # VTT parsing
β βββ ...
βββ background.ts # Service worker entry
βββ content.ts # Content script
βββ interceptor.ts # Caption interception
βββ manifest.ts # Extension manifest generator
CapSummarize includes 15 built-in summary styles:
- Default - Balanced, comprehensive summary
- Educational - Academic-focused with key concepts
- Technical - Code-focused with implementation details
- Casual - Friendly, conversational tone
- Executive - Business-focused brief
- Marketing - Engaging promotional style
- News - Journalistic article format
- Podcast - Audio-friendly notes
- Kids - Simple language for young audiences
- Blog - SEO-optimized blog post
- YouTube - Video description format
- Cheatsheet - Quick reference card
- Recap - Brief highlights
- Interview - Q&A format
- X/Twitter - Thread-ready format
You can also create custom variants in Settings.
Generate AI images by analyzing your video transcript (requires ChatGPT, Gemini, or Grok):
- Thumbnail (General) - Eye-catching YouTube thumbnails
- Thumbnail (MrBeast) - Explosive, bright colors, high energy
- Thumbnail (Casey Neistat) - Cinematic, documentary feel
- Thumbnail (Theo) - Tech-focused, clean modern design
- Thumbnail (5-Min Crafts) - DIY aesthetic, step-by-step
- Thumbnail (Linus Tech) - Tech review, product-focused
- Infographic - Visual data representation
- Comic Strip - Comic-style visual story
- Mind Map - Organic tree-style concept map
- Whiteboard - Hand-drawn diagram aesthetic
- Quote Card - Shareable quote with key insight
- Scene - Key moment visualization
Generate AI video clips by analyzing your transcript (Gemini only):
- Advertisement - Short promotional ad clip
- Trailer - Cinematic teaser trailer
- Quick Recap - Fast-paced summary clip
- Explainer - Educational explainer clip
- Cinematic - Artistic cinematic sequence
- Social Media - Viral social media clip
- No external servers - Everything runs locally in your browser
- No data collection - Your video content never leaves your device
- No accounts required - Use immediately without signup
- Open source - Audit the code yourself
See PRIVACY_POLICY.md and SECURITY_PRACTICES.md for details.
- YouTube
- Google Drive
- Google Sites
- Udemy
- Zoom
- X (Twitter)
- And more via caption interception
We welcome contributions! Please see CONTRIBUTING.md for guidelines.
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Make your changes
- Run lint and type checks (
npm run lint && npm run type-check) - 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.
- Built with TypeScript, Tailwind CSS, and Bun
- Icons from various AI providers
Made with β€οΈ by SamuxBuilds and the CapSummarize community















