hotjar

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs the agent to use Membrane actions (e.g., action list / action run with a Hotjar connection) to fetch Hotjar "Recordings", "Feedback", and "Surveys" — user-generated third-party content the agent is expected to read and that could materially influence decisions or tool use.