adversarial-ux-test

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s required workflow instructs the agent to “Run an adversarial UX test on [URL]”, then “Navigate to the app using browser tools” and “Attempt the persona’s ACTUAL TASKS” while reading the pages/content of that user-supplied URL at runtime, which is outsider-authored free text.