reviewstudio

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md shows it runs Membrane actions and proxy requests against Review Studio (e.g., "membrane action run" and "membrane request CONNECTION_ID /path/to/endpoint"), which pulls user-generated documents, annotations, and comments from a third-party service that the agent is expected to read and could use to drive subsequent actions.