chat-room

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path is the chat-room actor’s sendMessage(sender, text) action, where text is free-form client-authored input (outsider message content) that is then stored/broadcast as readable prose (message.text) into the actor’s context and event payloads.