An AI security platform that screens LLM application traffic for prompt injection, jailbreaks and sensitive-data leakage, aimed at giving security teams a policy layer over how employees and applications use AI models.