📝 Text to JSON Prompt Converter

A complete Full-Stack AI-powered application that converts plain text into context-aware, structured JSON prompts for better AI model comprehension.
The system features a React + Tailwind frontend for easy text input and history tracking, and a FastAPI + Pydantic AI + Mistral AI backend for intelligent JSON generation and validation.

🌐 Live Demo

• Frontend (Static Site): https://json-prompt-generator.onrender.com
• Backend API: https://json-prompt-backend.onrender.com

---

🚀 Features

🔹 Backend (FastAPI + Pydantic AI + Mistral AI)

• Advanced Pydantic AI Integration: Strong validation and structured responses
• Mistral AI Powered: Context-aware JSON generation with relevant keys only
• Smart Validation: Multi-layer type checking and error recovery
• Async Operations: High-performance async/await for scalability
• Robust Error Handling: Built-in retry logic, graceful fallbacks
• Multi-Domain Support: Handles code generation, travel planning, data science, writing, and more

🔹 Frontend (React + Tailwind)

• Instant Text-to-JSON Conversion
• Conversion History: Stores latest 5 conversions
• Copy to Clipboard for generated JSON
• Responsive UI with sidebar navigation
• Customizable JSON Structure

---

🛠 Tech Stack

Backend:

• FastAPI (Python 3.9+)
• Pydantic AI with Mistral AI integration
• Uvicorn
• python-dotenv

Frontend:

• React + TypeScript
• Vite
• Tailwind CSS
• Lucide React Icons

---

📂 Project Structure

JSON-Prompt-Generator/
├── backend/                  # FastAPI backend
│   ├── main.py
│   ├── services/
│   │   └── generator.py
│   ├── requirements.txt
│   └── .env
│
├── frontend/                  # React frontend
│   ├── src/
│   │   ├── App.tsx
│   │   ├── main.tsx
│   │   ├── components/
│   │   │   ├── Sidebar.tsx
│   │   │   ├── TextToJsonConverter.tsx
│   │   │   └── HistoryView.tsx
│   ├── index.html
│   ├── package.json
│   ├── tailwind.config.js
│   └── vite.config.ts
└── README.md

---

⚙️ Installation & Setup

1️⃣ Clone Repository

git clone https://github.com/MSG-Mutual-Support-to-Grow/JSON-Prompt-Generator.git
cd JSON-Prompt-Generator

---

2️⃣ Backend Setup

The backend setup instructions are in the backend/README.md

---

3️⃣ Frontend Setup

The frontend setup instructions are in the frontend/README.md

---

📡 API Endpoints

POST /prompts/transform

Transform plain text into structured JSON with context-aware key selection.

Request:

{
  "text": "plan a trip to Tokyo in December",
  "num_keys": 5,
  "include_ai_content": true
}

Response Example:

{
  "response": {
    "text": "plan a trip to Tokyo in December",
    "city": "Tokyo",
    "country": "Japan",
    "month": "December",
    "activities": ["sightseeing", "food tours"]
  },
  "source": "mistral"
}

GET /

Health check endpoint.

---

🖥 Example Workflows

• Programming → Generates language, framework, libraries, etc.
• Travel → Generates destination, budget, activities, etc.
• Content Writing → Generates topic, tone, audience, word_count.

---

🎨 Frontend Features

• Text Input → JSON Output instantly via API
• Copy JSON to Clipboard
• View Conversion History
• Responsive Design for desktop & mobile

---

🔧 Development

• Code Style: PEP 8 (backend), TypeScript best practices (frontend)
• Strong Typing: Pydantic models for backend validation
• CORS Ready: Accepts requests from http://localhost:3000 and http://localhost:5173

---

🐞 Troubleshooting

• Backend not starting? Activate venv and check .env file
• CORS issues? Update allow_origins in main.py
• Port conflicts? Change with --port flag in uvicorn or vite

---

📜 License

This project is licensed under the MIT License.

---

👥 Contributors

• aswathkarthick75-collab
• Joelrtharakan
• DicksonLegend
• gokul18062006
• Niranjan070
• snipergib
• hazim564-dotcom
• ThirupathiS-45
• Klassy01

🤝 Contributing

Pull requests are welcome! Please follow:

1. PEP 8 for Python code
2. TypeScript standards for frontend
3. Add docstrings/comments for clarity

---