mirror of
https://github.com/elder-plinius/LEAKHUB.git
synced 2026-02-12 16:52:53 +00:00
dcfb1b33dd66ea243e43fc7fc0688d69d8a08578
🚀 PRODUCTION READY: Complete platform overhaul with navigation, analytics, user management, and mobile optimization
🚀 PRODUCTION READY: Complete platform overhaul with navigation, analytics, user management, and mobile optimization
🚀 LeakHub - AI System Prompt Discovery Platform
The community hub for crowd-sourced system prompt leak verification. CL4R1T4S!
🌟 What is LeakHub?
LeakHub is a revolutionary web platform that enables the community to discover, submit, verify, and analyze AI system prompt leaks. It's designed to promote transparency in AI systems through crowdsourced verification and community-driven discovery.
✨ Features
🔍 Core Functionality
- Submit Leaks: Submit suspected AI system prompt leaks with detailed metadata
- Compare & Verify: Advanced comparison algorithms to verify authenticity
- Community Library: Browse and search through all submitted leaks
- Confidence Scoring: Automatic confidence assessment based on content analysis
🎯 Target Types Supported
- 🤖 AI Models (GPT-4, Claude, Gemini, etc.)
- 📱 Apps/Interfaces (Cursor, GitHub Copilot, etc.)
- 🔧 Tools/Functions (Code Interpreter, WebPilot, etc.)
- 🤝 AI Agents (AutoGPT, BabyAGI, etc.)
- 🔌 Plugins/Extensions (Browser extensions, IDE plugins, etc.)
- 🛠️ Custom GPTs/Bots (Custom implementations)
🏆 Gamification & Community
- Leaderboard: Track top contributors and achievements
- Daily Challenges: Compete in daily leak hunting challenges
- Achievement System: Unlock badges for notable contributions
- Points System: Earn points for submissions, verifications, and discoveries
🎯 Requests & Bounties
- Community Requests: Request specific targets for leak hunting
- Voting System: Vote on which targets to prioritize
- Bounty System: Offer points for high-priority discoveries
- Trending Requests: See what the community wants most
📊 Advanced Analytics
- Similarity Metrics: Character match, word match, structure match, core similarity
- Consensus View: Identify common elements across multiple submissions
- Verification Tracking: Track verification status and confidence levels
- Statistics Dashboard: Real-time platform statistics
🚀 Getting Started
Prerequisites
- For Development: Node.js 18+ and npm
- For Production: Modern web browser (Chrome, Firefox, Safari, Edge)
- For Backend: Optional - MongoDB, Supabase, or Firebase
Quick Start (Development)
# Clone the repository
git clone https://github.com/elder-plinius/LEAKHUB.git
cd LEAKHUB
# Install dependencies
npm install
# Start development server
npm run dev
# Open http://localhost:3000
Quick Start (Production)
- GitHub Pages: Push to main branch - auto-deploys!
- Vercel:
npm i -g vercel && vercel - Netlify:
npm install -g netlify-cli && netlify deploy
Local Testing
# Build for production
npm run build
# Preview production build
npm run preview
# Serve static files
npm run serve
Quick Start Guide
1. Submit Your First Leak
- Fill in your identifier (username)
- Select the target type (AI Model, App, Tool, etc.)
- Enter the target name and URL
- Paste the suspected system prompt
- Add context and access requirements
- Submit and earn points!
2. Compare Submissions
- Select two different submissions from the dropdowns
- Click "Compare" to analyze similarity
- Review the detailed metrics and consensus view
- High similarity automatically boosts confidence scores
3. Join the Community
- Click "View Leaderboard" to see top contributors
- Check out "Requests & Challenges" for daily missions
- Vote on community requests
- Track your achievements and progress
🎮 How It Works
Submission Process
- Content Analysis: Automatic confidence scoring based on prompt patterns
- First Discovery Detection: Identifies if this is the first submission for a target
- Point Allocation: Awards points for submissions, discoveries, and verifications
- Metadata Tracking: Stores access requirements, context, and tool information
Verification System
- Multi-Metric Comparison: Uses character, word, structure, and core similarity
- Consensus Building: Identifies common elements across submissions
- Confidence Boosting: High similarity automatically increases confidence scores
- Verification Milestones: Tracks when leaks reach verified status
Gamification Mechanics
- Base Points: 10 points per submission
- First Discovery: 100 points + 50 bonus for non-model targets
- Tool Prompts: 30 bonus points for comprehensive submissions
- Verification: 20 points per verification + 50 bonus for reaching verified status
- Daily Challenges: 400-700 points for completing daily missions
🎨 Design Features
Modern UI/UX
- Cyberpunk Aesthetic: Dark theme with neon accents and gradients
- Responsive Design: Works perfectly on desktop, tablet, and mobile
- Smooth Animations: Floating grid background and interactive elements
- Intuitive Navigation: Clear sections and easy-to-use interface
Visual Elements
- Status Indicators: Real-time statistics and progress tracking
- Badge System: Visual indicators for access requirements and features
- Color Coding: Different colors for different target types and confidence levels
- Interactive Elements: Hover effects, transitions, and feedback
🔧 Technical Details
Frontend Technologies
- HTML5: Semantic markup and modern structure
- CSS3: Advanced styling with animations and responsive design
- JavaScript (ES6+): Full application logic and data management
- Vite: Fast build tool and development server
- LocalStorage: Client-side data persistence with backend fallback
Backend & Deployment
- GitHub Actions: Automated CI/CD pipeline
- Vercel/Netlify: Serverless deployment with functions
- Database Abstraction: Support for localStorage, MongoDB, Supabase, Firebase
- API Layer: RESTful endpoints with CORS support
Key Algorithms
- Text Normalization: Standardized text processing for comparison
- Similarity Metrics: Multiple algorithms for comprehensive analysis
- Levenshtein Distance: String similarity calculation
- Phrase Matching: Sliding window approach for common phrase detection
Data Management
- Hybrid Storage: localStorage with optional backend persistence
- JSON Serialization: Efficient data storage and retrieval
- Real-time Updates: Instant UI updates and statistics
- Data Validation: Input validation and error handling
- Export/Import: Full data backup and restore functionality
🎯 Use Cases
For Researchers
- Document and verify AI system prompts
- Compare different implementations
- Track prompt evolution over time
- Build comprehensive prompt databases
For Developers
- Understand AI system behaviors
- Debug prompt-related issues
- Learn from existing implementations
- Contribute to AI transparency
For Community
- Participate in AI transparency efforts
- Compete in daily challenges
- Build reputation and achievements
- Help verify and validate discoveries
🤝 Contributing
How to Contribute
- Submit Leaks: Share discovered system prompts
- Verify Submissions: Compare and verify existing leaks
- Request Targets: Suggest new targets for the community
- Vote on Requests: Help prioritize community goals
- Report Issues: Help improve the platform
Best Practices
- Be Accurate: Only submit genuine system prompts
- Provide Context: Include how you obtained the prompt
- Add Metadata: Include access requirements and URLs
- Verify Others: Help verify community submissions
- Respect Privacy: Don't submit private or sensitive information
🔮 Future Enhancements
Planned Features
- Advanced Analytics: Machine learning-based similarity detection
- Mobile App: Native mobile application
- Real-time Collaboration: Live collaboration features
- Advanced Export: Data export in various formats (CSV, JSON, API)
- User Authentication: Secure user accounts and profiles
- Social Features: Comments, discussions, and community forums
Backend Integration
- Database Support: MongoDB, PostgreSQL, Firebase integration
- API Access: RESTful API for programmatic access
- Real-time Updates: WebSocket support for live updates
- Email Notifications: Automated notifications and alerts
- Advanced Security: JWT authentication and rate limiting
Community Features
- User Profiles: Detailed user profiles and statistics
- Discussion Forums: Community discussions and analysis
- Expert Verification: Expert review system
- Bounty Marketplace: Advanced bounty and reward system
- Achievement System: Expanded badges and rewards
📄 License
This project is open source and available under the MIT License. See the LICENSE file for details.
🙏 Acknowledgments
- AI Research Community: For inspiration and feedback
- Open Source Contributors: For tools and libraries used
- Early Testers: For valuable feedback and suggestions
- CL4R1T4S Community: For the vision of AI transparency
📞 Support
- Issues: Report bugs and feature requests via GitHub issues
- Discussions: Join community discussions
- Documentation: Check the inline documentation and comments
Ready to start hunting for AI system prompts? Open index.html and join the LeakHub community! 🚀
Description
Languages
TypeScript
93.8%
JavaScript
3.8%
CSS
2.1%
HTML
0.3%