Guardrails are emerging as a critical piece of LLM app architecture. These tools can help block jailbreak attempts, harmful responses, and malicious behavior in your systems.
Guardrails AI (https://www.guardrailsai.com/) is a great tool to help you manage these guards. Our new integration allows you to view trace data on your guards and create datasets of triggered inputs and outputs to hone in on potential attack vectors.
We've even created our own guard in the Guardrails ecosystem to help you keep your apps secure.
Check out the full integration with our open source platform Phoenix here, and look out for a follow up on what this looks like in Arize!
📓 Notebook: https://colab.research.google.com/drive/1NDn5jzsW5k0UrwaBjZenRX29l6ocrZ-_?usp=sharing
⭐️ Star Phoenix on GitHub: https://github.com/Arize-ai/phoenix
