DigitVision v1.0.x is the only currently supported version.

DigitVision is a portfolio/educational application that runs locally or in a Docker container on a private machine. It does not:

• Accept arbitrary code execution from external users
• Store user credentials or personal data
• Connect to external APIs or databases
• Expose a public-facing network endpoint by default

The primary security concerns for this project are:

1. Dependency vulnerabilities in tensorflow, streamlit, pillow, or opencv-python-headless
2. Container hardening (the Dockerfile runs as a non-root user)
3. Model file integrity (trained .keras files loaded from disk)

If you discover a security vulnerability, please do not open a public GitHub issue. Instead, report it privately:

Email: [your-email@example.com]

Please include:

• A description of the vulnerability
• Steps to reproduce
• The potential impact
• Any suggested mitigations

You will receive a response within 72 hours. If the vulnerability is confirmed, a patch will be prepared and a security advisory published before public disclosure (coordinated disclosure).

You can audit the current dependency tree for known vulnerabilities using:

pip install pip-audit
pip-audit -r requirements.txt

Or with the Safety CLI:

pip install safety
safety check -r requirements.txt

The production Dockerfile:

• Uses python:3.12-slim (minimal attack surface)
• Runs as a non-root user (digitvision)
• Does not expose any port other than 8501
• Uses --no-install-recommends for all apt packages

When deploying publicly, place the app behind a reverse proxy (nginx, Caddy) with TLS and appropriate firewall rules.