rocket-chat

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly instructs running Membrane actions and proxy requests against a Rocket.Chat connection (e.g., "membrane action run" and "membrane request CONNECTION_ID /path/to/endpoint"), which will fetch user-generated messages/rooms/users from Rocket.Chat that the agent is expected to read and could contain instructions influencing subsequent actions.