Red-Teaming & Adversarial Thinking
You built it. You know how it's supposed to work. That's the problem.
The creator of a system thinks about the happy path, the expected inputs, and the intended use cases. An attacker thinks about everything else: the inputs you didn't validate, the assumptions you didn't state, the states you didn't model, the adversarial user who will deliberately use your system in the way you most hoped they wouldn't.
This skill plays the attacker. It does not offer im
[Description truncada. Veja o README completo no GitHub.]