ZChat

A privacy-preserving AI-powered chat application with cryptocurrency integration, built using NEAR AI Cloud that runs inside a Trusted Execution Environment (TEE). ZChat enables users to interact with an intelligent assistant that can fetch crypto prices, provide charity information, and facilitate hotel bookings - all while maintaining privacy through TEE.

🌟 Features

• Privacy-Preserving AI: Powered by NEAR AI Cloud running inside TEE (Trusted Execution Environment) for secure, private AI computations
• Wallet Integration: Connect your Solana wallet for seamless Web3 interactions
• Cryptocurrency Price Tracking: Real-time crypto prices with interactive charts powered by Plotly
• Charity Donations: Browse and donate to children's cancer research charities
• Hotel Bookings: Search and book hotels with cryptocurrency
• Chat Persistence: MongoDB-backed chat history storage
• Model Context Protocol (MCP): Tool-calling capabilities for extended functionality
• Modern UI: Chat interface with markdown support and real-time updates

🏗️ Architecture

User Input
    ↓
ChatInterface (React/Next.js Frontend)
    ↓
POST /api/chat (Next.js API Route)
    ↓
NEAR AI Cloud (DeepSeek V3.1)
    ↓
MCP Client ←→ MCP Server (Tools)
    ↓
    ├─ CoinGecko API (Price Data)
    ├─ Charities JSON (Mock Data)
    ├─ Hotels JSON (Mock Data)
    └─ MongoDB (Chat Persistence)

🛠️ Tech Stack

Frontend

• Next.js 16 - React framework with App Router
• React 19 - UI library
• TypeScript - Type safety
• Tailwind CSS - Styling
• Plotly.js - Interactive cryptocurrency charts
• React Markdown - Rich message formatting

Backend

• Next.js API Routes - Serverless API endpoints
• NEAR AI Cloud - Privacy-preserving LLM with TEE
• MongoDB - Chat history and wallet storage
• Model Context Protocol (MCP) - Tool integration framework

MCP Server Tools

• get_token_price - Fetch crypto prices from CoinGecko
• get_charity_details - Retrieve charity information
• get_hotel_details - Search and retrieve hotel data
• process_payment - Handle cryptocurrency transactions
• get_price_history - Fetch historical price data for charts

📋 Prerequisites

Before setting up the project, ensure you have:

• Node.js (v18 or higher)
• pnpm (recommended package manager)
• MongoDB instance (local or cloud)
• NEAR AI Cloud API key (Get one here)

🚀 Setup Instructions

1. Clone the Repository

git clone <repository-url>
cd zchat

2. Install Dependencies

pnpm install

3. Configure Environment Variables

Create a .env file in the root directory with the following variables:

# NEAR AI Cloud Configuration
NEAR_API_KEY=your_near_ai_cloud_api_key_here

# Site Configuration
SITE_URL=http://localhost:3000

# MongoDB Configuration
MONGODB_URI=mongodb://localhost:27017/Zchat
# or use MongoDB Atlas:
# MONGODB_URI=mongodb+srv://<username>:<password>@cluster.mongodb.net/Zchat

# Database and Collection Names
DB_NAME=Zchat
COLLECTION_NAME=chat

Getting Your NEAR AI Cloud API Key:

1. Visit https://cloud.near.ai/
2. Sign up or log in to your account
3. Navigate to API Keys section
4. Generate a new API key
5. Copy and paste it into your .env file

4. Set Up MongoDB

Option A: Local MongoDB

# Install MongoDB (Ubuntu/Debian)
sudo apt-get install mongodb

# Start MongoDB service
sudo systemctl start mongodb

Option B: MongoDB Atlas (Cloud)

1. Create a free account at MongoDB Atlas
2. Create a new cluster
3. Get your connection string
4. Add it to your .env file as MONGODB_URI

5. Run the Development Server

# Start the Next.js development server
pnpm dev

The application will be available at http://localhost:3000

6. MCP Server

The MCP server runs automatically when the API is called. However, you can test it independently:

# Run MCP server standalone
pnpm run mcp-server

📖 Usage

Example Prompts to Try

Once the application is running, try these prompts in the chat:

1. Cryptocurrency Prices:

   • "What's the current price of Bitcoin?"
   • "Show me the price of ZEC in USD"
   • "Get me the price history of Ethereum"

2. Charity Donations:

   • "Show me charities for children's cancer research"
   • "Tell me about charity #1"
   • "I want to donate to a cancer research charity"

3. Hotel Bookings:

   • "Find hotels in Miami"
   • "Show me hotels in Dubai"
   • "Tell me about the rooms at hotel #2"

4. Interactive Charts:

   • "Show me Bitcoin price history for the last 7 days"
   • "Display Ethereum price chart"

🔧 Project Structure

zchat/
├── app/
│   ├── api/
│   │   └── chat/
│   │       └── route.ts          # Main chat API endpoint
│   ├── components/
│   │   ├── ChatInterface.tsx     # Main chat UI component
│   │   ├── MarkdownRenderer.tsx  # Message rendering with markdown
│   │   ├── PriceChart.tsx        # Plotly chart component
│   │   └── ...
│   ├── globals.css               # Global styles
│   ├── layout.tsx                # Root layout
│   └── page.tsx                  # Home page
├── data/
│   ├── charities.json            # Charity data (mocked)
│   └── hotels.json               # Hotel data (mocked)
├── lib/
│   └── mongodb.ts                # MongoDB connection and utilities
├── mcp-server/
│   ├── index.ts                  # MCP server implementation
│   └── README.md                 # MCP server documentation
├── public/                       # Static assets
├── .env                          # Environment variables (create this)
├── package.json                  # Project dependencies
└── README.md                     # This file

🔐 Privacy & Security

ZChat uses NEAR AI Cloud which runs inside a Trusted Execution Environment (TEE). This means:

• All AI computations happen in an isolated, secure environment
• Your conversations are processed privately and securely
• Data is encrypted during processing
• No unauthorized access to your chat data or wallet information

📝 Important Notes

Mocked Data for Testing

This project includes mocked data for hotels and charities in the data/ directory:

• data/hotels.json - Contains mock hotel listings with various room types and pricing
• data/charities.json - Contains mock children's cancer research charities

Why Mocked Data?

These charities and hotels don't allow testing with test wallets on testnet environments. To provide a seamless development and testing experience without requiring real transactions, we've created realistic mock data that simulates the actual behavior of these services. This allows you to:

• Test the full user flow without real payments
• Develop and debug features safely
• Demonstrate the application's capabilities without financial risk

For production deployment, these would need to be replaced with real API integrations and proper payment processing systems that support mainnet transactions.

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is open source and available under the MIT License.

🆘 Troubleshooting

MongoDB Connection Issues

# Check if MongoDB is running
sudo systemctl status mongodb

# Restart MongoDB
sudo systemctl restart mongodb

Dependencies Installation Issues

# Clear cache and reinstall
rm -rf node_modules pnpm-lock.yaml
pnpm install

NEAR API Connection Issues

• Verify your API key is correct in .env
• Check if you have an active internet connection
• Ensure your API key has not expired

MCP Server Issues

• Make sure all dependencies are installed
• Check that TypeScript compilation is working: pnpm run mcp-server
• Review MCP server logs for specific errors

📚 Additional Resources

• NEAR AI Cloud Documentation
• Model Context Protocol
• Next.js Documentation
• MongoDB Documentation
• CoinGecko API

---

Built with ❤️ using NEAR AI Cloud and privacy-preserving technology