Skip to content

MeghvShetty/model-armor-redteam

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Model Armor Red Team

An adversarial testing framework for Google Cloud's Model Armor AI firewall. Deploys Model Armor via Terraform and runs an autonomous ADK-based attack agent across five control categories: prompt injection, jailbreak, hate speech, malicious URL detection, and sensitive data protection.

Built to answer the question: What does a GCP AI content filter actually stop?

About

An adversarial AI security testing framework that deploys a GCP AI firewall via IaC, then runs an autonomous red team agent against it across 5 attack categories.

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors