O9 Automation Platform

An AI-powered test automation platform for O9 Supply Chain Planning, built for Mondelez. This system allows non-technical supply chain users to create, manage, and execute automated tests using natural language.

Features

• Natural Language Test Generation: Convert plain English descriptions into automated test scripts
• AI-Powered Script Creation: Uses Claude API to generate Selenium automation scripts
• Inline Editing: Edit test steps directly and regenerate scripts
• Real-time Execution: Watch tests run with live browser screenshots
• Mock O9 Environment: Test automation without access to production O9

Tech Stack

Backend

• Python 3.12
• FastAPI
• SQLite (development) / PostgreSQL (production)
• Anthropic Claude Sonnet 4 API
• Selenium WebDriver
• WebSocket (FastAPI WebSocket)

Frontend

• React + Vite
• Tailwind CSS
• Socket.IO Client

Project Structure

O9Automation/
├── backend/
│   ├── app/
│   │   ├── __init__.py
│   │   ├── routes.py
│   │   ├── models.py
│   │   ├── database.py
│   │   ├── websocket_handler.py
│   │   └── services/
│   │       ├── ai_service.py
│   │       ├── ai_selenium_generator.py
│   │       ├── selenium_executor.py
│   │       └── selenium_service.py
│   ├── instance/
│   │   └── test_cases.db
│   ├── requirements.txt
│   ├── .env
│   └── run.py
├── frontend/
│   ├── src/
│   │   ├── components/
│   │   │   ├── TestStepExecutor.jsx
│   │   │   ├── TestCaseDetail.jsx
│   │   │   └── ...
│   │   ├── App.jsx
│   │   └── main.jsx
│   ├── package.json
│   └── vite.config.js
├── mock-o9-website/
│   ├── index.html
│   ├── dashboard.html
│   ├── forecast.html
│   └── css/
├── .gitignore
└── README.md

Setup Instructions

Backend Setup

1. Navigate to backend directory:

   cd backend

2. Create virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt --break-system-packages

4. Create .env file:

   # Create backend/.env file with:
   ANTHROPIC_API_KEY=your_api_key_here
   O9_MOCK_URL=http://localhost:3001

5. Initialize database:

   python run.py
   # Database will be created automatically in instance/ folder

6. Run backend:

   python run.py
   # Server will start on http://localhost:8000

Frontend Setup

1. Navigate to frontend directory:

   cd frontend

2. Install dependencies:

   npm install

3. Run frontend:

   npm run dev
   # App will start on http://localhost:5173

Mock O9 Website Setup (for testing)

1. Navigate to mock website directory:

   cd mock-o9-website

2. Serve the website:

   # Option 1: Python
   python3 -m http.server 3001
   
   # Option 2: Node.js
   npx http-server -p 3001

3. Access at: http://localhost:3001

Usage

1. Create Test Case: Navigate to the dashboard and create a new test case
2. Add Test Steps: Describe each test step in natural language
3. Generate Scripts: Click "Generate Script" to create automation
4. Edit Steps: Click "Edit Step" to modify descriptions
5. Regenerate: Click "Regenerate Script" after editing
6. Run Tests: Click "Run Step" to execute automation
7. View Results: See execution status and screenshots in real-time

API Endpoints

Test Cases

• GET /api/test-cases - List all test cases
• POST /api/test-cases - Create new test case
• GET /api/test-cases/<id> - Get test case details
• PUT /api/test-cases/<id> - Update test case
• DELETE /api/test-cases/<id> - Delete test case

Test Steps

• GET /api/test-cases/<id>/steps - Get all steps for a test case
• POST /api/test-cases/<id>/steps - Create new step
• PUT /api/test-cases/<id>/steps/<step_id> - Update step
• DELETE /api/test-cases/<id>/steps/<step_id> - Delete step
• POST /api/test-steps/<id>/generate-selenium - Generate automation script

WebSocket Events

• execute_step - Execute a single test step
• status_update - Real-time execution status
• progress - Execution progress updates
• screenshot - Live browser screenshots
• execution_complete - Step execution finished

Architecture

Test Generation Flow

User Input (Natural Language)
    ↓
Claude API (Anthropic)
    ↓
JSON Commands (for execution) + Python Script (for display)
    ↓
Selenium WebDriver
    ↓
Browser Automation
    ↓
Results + Screenshots

Execution Flow

User clicks "Run Step"
    ↓
WebSocket event triggered
    ↓
Backend initializes Chrome browser
    ↓
Executes JSON commands sequentially
    ↓
Sends real-time updates (progress, screenshots)
    ↓
Returns final status (passed/failed/error)
    ↓
Saves results to database

Configuration

Backend (.env)

ANTHROPIC_API_KEY=your_key_here
O9_MOCK_URL=http://localhost:3001
DATABASE_URL=sqlite:///instance/test_cases.db

Frontend

• API URL: http://localhost:8000
• WebSocket: ws://localhost:8000

Development Notes

• ChromeDriver: Managed automatically by Selenium Manager (4.6+)
• Database: SQLite for development, PostgreSQL for production
• AI Model: Claude Sonnet 4 (claude-sonnet-4-20250514)
• Browser: Headless Chrome for automation
• JSON-Only Execution: System only executes JSON commands, never Python scripts

Known Issues

• ChromeDriver may require manual installation on some systems
• WebSocket connections need CORS configuration for different origins
• Database migrations not automated (manual schema updates needed)

Future Enhancements

• Production deployment configuration
• PostgreSQL migration
• User authentication and multi-tenancy
• Test scheduling and CI/CD integration
• Advanced reporting and analytics
• Element locator recorder tool
• Actual O9 integration (replacing mock)

License

Proprietary - Mondelez International

Contact

For questions or support, contact the development team.

---

Note: This is an internal tool for Mondelez. Do not share API keys or sensitive information.