shadrackotina-lab/kenya-ai-sales

Project: kenya-ai-sales # AI WHATSAPP + M-PESA SALES SYSTEM This project is a production-grade SaaS web application for Kenyan businesses to automate sales using AI and M-Pesa. ## Tech Stack - **Frontend**: React 19 + Vite (UI mapped from requested Next.js design) - **Styling**: Tailwind CSS + Shadcn UI - **Animations**: Framer Motion - **Icons**: Lucide React - **Charts**: Recharts - **Logic**: Daraja M-Pesa API integration, OpenAI GPT models ## Setup Instructions 1. **Install Dependencies**: ```bash bun install ``` 2. **Configure Environment Variables**: Create a `.env` file with: ```env OPENAI_API_KEY=your_key MPESA_CONSUMER_KEY=your_key MPESA_CONSUMER_SECRET=your_secret MPESA_SHORTCODE=174379 MPESA_PASSKEY=bfb279f9aa9bdbcf158e97ddf5fddf5fddf5fddf5fddf5fddf5fddf5fdd BASE_URL=localhost ``` 3. **Run Development Server**: ```bash bun dev ``` ## M-Pesa Sandbox Testing - Use the sandbox credentials provided in the code. - Test phone number: `254708374149`. - Amount: `1`. - The callback URL must be reachable (use Ngrok for local testing). ## Architecture - `src/App.tsx`: Main router. - `src/pages/LandingPage.tsx`: Marketing site with interactive AI demo. - `src/pages/Dashboard.tsx`: Sales and analytics metrics. - `src/lib/mpesa/`: Core Daraja API logic. - `src/lib/openai.ts`: AI assistant orchestration.