A web application that generates credit ratings for public companies using LLM-based methodology interpretation and Box platform integration.

• Generate credit ratings for public companies
• LLM-based methodology interpretation from PDF documents
• Integration with Box platform for document management
• Historical ratings tracking
• RESTful API with interactive documentation

• Python 3.11 or higher
• PostgreSQL database
• Box Custom Application (JWT authentication)
• OpenAI API key
• Financial data provider API key (e.g., Alpha Vantage)

1. Clone the repository:

git clone <repository-url>
cd company-credit-rating

2. Create and activate a virtual environment:

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

3. Install dependencies:

pip install -e ".[dev]"

4. Set up environment variables:

cp .env.example .env
# Edit .env with your configuration

5. Set up PostgreSQL database:

createdb credit_rating_db

6. Initialize the database:

python -c "from app.database import init_db; init_db()"

The application uses the self-hosted Box MCP server for Box integration. Follow these steps:

1. Install the Box MCP server following the official guide: https://developer.box.com/guides/box-mcp/self-hosted/

2. Configure your Box Custom Application:

   • Create a Custom App in Box Developer Console
   • Enable JWT authentication
   • Generate and download the configuration file
   • Grant necessary permissions (read/write files, manage folders)

3. Update .env with Box configuration:

   • BOX_MCP_SERVER_PATH: Path to the Box MCP server executable
   • BOX_CLIENT_ID, BOX_CLIENT_SECRET, BOX_ENTERPRISE_ID: From Box app configuration
   • BOX_METHODOLOGY_FOLDER_ID: Box folder ID containing methodology PDFs
   • BOX_RATINGS_FOLDER_ID: Box folder ID for storing rating reports

python app/main.py

The application will be available at http://localhost:5000

API documentation is available at http://localhost:5000/api/docs

export FLASK_APP=app.main:create_app
export FLASK_ENV=development
flask run

pytest

pytest --cov=app --cov=models --cov=services --cov-report=html

pytest tests/unit/

pytest tests/integration/ -v -s

# Using the helper script
./scripts/test_sec_to_box.sh

# Or directly with pytest
pytest tests/integration/test_sec_to_box_e2e.py -v -s

pytest tests/properties/

See tests/integration/README.md for detailed information about integration tests.

company-credit-rating/
├── app/                    # Flask application
│   ├── __init__.py
│   ├── main.py            # Application factory
│   ├── config.py          # Configuration
│   ├── database.py        # Database setup
│   └── routes/            # API endpoints (to be created)
├── models/                # Data models
│   └── __init__.py
├── services/              # Business logic services
│   └── __init__.py
├── tests/                 # Test suite
│   ├── unit/             # Unit tests
│   ├── properties/       # Property-based tests
│   └── conftest.py       # Test configuration
├── .env.example          # Environment variables template
├── .gitignore
├── pyproject.toml        # Project configuration
└── README.md

Key configuration options in .env:

• Flask: SECRET_KEY, FLASK_ENV
• Database: DATABASE_URL
• Box: BOX_MCP_SERVER_PATH, BOX_CLIENT_ID, etc.
• LLM: OPENAI_API_KEY, LLM_MODEL, LLM_TEMPERATURE
• Financial Data: FINANCIAL_DATA_PROVIDER, ALPHA_VANTAGE_API_KEY
• Application: MAX_RETRIES, REQUEST_TIMEOUT, DATA_AGE_THRESHOLD_MONTHS

black .

mypy app/ models/ services/

flake8 app/ models/ services/

[Your License Here]