We provide a first line of defense against prompt injection attacks

Our team is constantly researching prompt injection attack vectors and developing and curating powerful rulesets to detect and block them.