Introducing Augustus: LLM prompt-injection testing

Augustus tests large-language-model applications for prompt injection and related model-abuse paths.

It turns those tests into Guard findings so AI features are tracked with the rest of the application risk.