AutoDeploy - Automated App Generator API

A FastAPI application that automatically generates web applications based on user requirements, creates GitHub repositories, and deploys them to GitHub Pages.

Features

• AI-powered code generation using OpenAI
• GitHub repository creation and management
• Automatic GitHub Pages deployment
• Webhook notifications on completion
• Secure authentication with secrets
• Support for file attachments
• Duplicate request detection

Local Development

Prerequisites

• Python 3.13.7
• GitHub account and token
• OpenAI API key

Setup

1. Create a virtual environment:

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

2. Install dependencies:

   pip install -r requirements.txt

3. Configure environment variables: Create a .env file with:

   GITHUB_TOKEN=your_github_token
   GITHUB_USERNAME=your_github_username
   OPENAI_API_KEY=your_openai_api_key
   USER_SECRET=your_secret_key
   OPENAI_BASE_URL=https://aipipe.org/openai/v1

4. Run the application:

   uvicorn app.main:app --reload

The API will be available at http://127.0.0.1:8000

API Documentation

Once running, visit:

• Interactive API docs: http://127.0.0.1:8000/docs
• Alternative docs: http://127.0.0.1:8000/redoc

API Endpoint

POST /api-endpoint

Generate and deploy a web application.

Request Body:

{
  "email": "[email protected]",
  "secret": "your_secret",
  "task": "project-name",
  "round": 1,
  "nonce": "unique-identifier",
  "brief": "Description of the app to generate",
  "checks": ["Requirement 1", "Requirement 2"],
  "evaluation_url": "https://webhook.site/your-id",
  "attachments": []
}

Example:

curl -X POST http://127.0.0.1:8000/api-endpoint \
-H "Content-Type: application/json" \
-d '{
  "email": "[email protected]",
  "secret": "tdsproject1_2025",
  "task": "weather-app",
  "round": 1,
  "nonce": "test-001",
  "brief": "Create a weather app with city search",
  "checks": ["Has README", "Works on mobile"],
  "evaluation_url": "https://webhook.site/your-id",
  "attachments": []
}'

Deployment

See DEPLOYMENT.md for detailed instructions on deploying to Render.

Quick Deploy to Render

1. Push code to GitHub
2. Connect repository to Render
3. Add environment variables
4. Deploy!

Your API will be available at https://your-app.onrender.com

Project Structure

tds-project-1-main/
├── app/
│   ├── __init__.py
│   ├── main.py              # FastAPI application
│   ├── llm_generator.py     # OpenAI integration
│   ├── github_utils.py      # GitHub API functions
│   ├── notify.py            # Webhook notifications
│   └── signature.py         # Request signing
├── .env                     # Environment variables (not in git)
├── .gitignore              # Git ignore rules
├── requirements.txt        # Python dependencies
├── runtime.txt            # Python version
├── Procfile               # Process file for deployment
├── build.sh               # Build script for Render
├── render.yaml            # Render configuration
└── README.md              # This file

How It Works

1. Receives a request with project requirements
2. Uses OpenAI to generate code based on the brief
3. Creates a GitHub repository
4. Commits generated files (HTML, README, LICENSE)
5. Enables GitHub Pages
6. Sends completion notification to evaluation URL

Development

Running Tests

python test_github.py

Checking Logs

The application logs to stdout. In production, check Render logs.

Environment Variables

Variable	Description	Required
GITHUB_TOKEN	GitHub personal access token	Yes
GITHUB_USERNAME	Your GitHub username	Yes
OPENAI_API_KEY	OpenAI API key	Yes
USER_SECRET	Secret for request authentication	Yes
OPENAI_BASE_URL	OpenAI API base URL	No

License

MIT License - See LICENSE file for details

Support

For issues or questions:

• Check the DEPLOYMENT.md guide
• Review API documentation at /docs
• Check application logs