I built an open-source Python package that scans LLM inputs and outputs for injections (opens in new tab)

Discussed on r/Python

AI security guardrails for LLM applications. Contribute to Adxzer/defend development by creating an account on GitHub.