Hook Forge is now fully operational! This is an AI-powered SaaS platform for X (Twitter) content creators who want to write better hooks.
Content creators struggle to write compelling opening hooks. Most don't know what makes a hook effective or how to analyze why some posts go viral. Hook Forge uses AI to solve this.
-
Hook Generator
- Enter any topic → Get 5 AI-generated hook variations
- Powered by OpenAI GPT-4
- All hooks optimized for X's 280-character limit
- Copy hooks with one click
-
Hook Analyzer
- Paste any hook → Get instant score (0-100)
- Detailed feedback on:
- Clarity and simplicity
- Emotional impact
- Curiosity factor
- Specificity
- Specific improvement suggestions
-
Pattern Library
- 10 proven hook templates from viral posts
- Categories: Questions, Lists, Bold Statements, Stories, Data, etc.
- Real examples for each pattern
- Copy templates directly
- Filter by category
- Frontend: React 18 + TypeScript + Vite + Tailwind CSS
- Backend: Node.js + Express + TypeScript
- Database: JSON file-based (simple, no build tools required)
- AI: OpenAI GPT-4 API
- Development: Concurrently runs both servers
hook-forge/
├── client/ # React frontend (port 5173)
│ ├── src/
│ │ ├── components/
│ │ │ ├── Generator.tsx # Hook generation UI
│ │ │ ├── Analyzer.tsx # Hook analysis UI
│ │ │ └── Patterns.tsx # Pattern library UI
│ │ ├── App.tsx # Main app with tab navigation
│ │ ├── api.ts # API client (axios)
│ │ └── main.tsx # Entry point
│ ├── package.json
│ └── vite.config.ts
│
├── server/ # Express backend (port 3001)
│ ├── src/
│ │ ├── routes/
│ │ │ └── hooks.ts # API endpoints
│ │ ├── services/
│ │ │ └── ai.ts # OpenAI integration
│ │ ├── db.ts # JSON database
│ │ └── index.ts # Server entry
│ ├── data/
│ │ └── db.json # Data storage (auto-created)
│ ├── package.json
│ └── tsconfig.json
│
├── package.json # Root package with scripts
├── README.md # Setup & usage docs
├── AGENTS.md # Developer guide
└── .gitignore
- Frontend: http://localhost:5173
- Backend: http://localhost:3001
- Health Check: http://localhost:3001/health ✅
Both servers are running in watch mode with hot reload enabled.
npm run dev # Run both frontend & backend
npm run dev:server # Run only backend
npm run dev:client # Run only frontend
npm run build # Build for production
npm start # Start production server
npm run install:all # Install all dependencies| Method | Endpoint | Description |
|---|---|---|
| POST | /api/hooks/generate |
Generate hooks for a topic |
| POST | /api/hooks/analyze |
Analyze and score a hook |
| GET | /api/hooks/patterns |
Get all hook patterns |
| GET | /api/hooks/history |
Get hook generation history |
| GET | /health |
Server health check |
- Modern gradient design (blue to purple)
- Tab-based navigation
- Real-time character count
- Copy-to-clipboard functionality
- Color-coded scoring (green/yellow/red)
- Responsive layout
- Loading states & error handling
The app requires an OpenAI API key to function:
- Get API key from: https://platform.openai.com/api-keys
- Add to
server/.env:OPENAI_API_KEY=sk-your-key-here
Uses a simple JSON file (server/data/db.json) that stores:
- Generated hooks with timestamps
- Hook analysis results
- 10 pre-seeded hook patterns
Benefits:
- No build tools needed (unlike SQLite)
- Easy to inspect and debug
- Automatically created on first run
- Simple backup (just copy the file)
-
Authentication & User Accounts
- Sign up / login
- Personal hook library
- Save favorite patterns
-
Analytics Dashboard
- Track hook performance
- A/B testing
- Engagement metrics
-
Advanced Features
- Hook variations (rewrite in different styles)
- Competitor analysis
- Trending hook patterns
- Chrome extension for X.com
-
Monetization
- Free tier: 10 hooks/day
- Pro tier: Unlimited + analytics
- Stripe integration
-
Collaboration
- Team workspaces
- Share hook templates
- Commenting & feedback
- Switched from SQLite to JSON database to avoid Windows build tool issues
- Frontend on port 5173 (Vite default) to avoid port conflicts
- Backend on port 3001 with CORS enabled
- Vite proxy forwards
/api/*to backend - GPT-4 used for both generation (temp: 0.9) and analysis (temp: 0.7)
None at this time! App is fully functional.
- README.md: Installation and usage guide
- AGENTS.md: Architecture and development guide for future AI agents
- Code is well-organized and commented