happyfox-chat

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs using Membrane actions such as "list-transcripts" and "get-transcript" (and the Membrane proxy via membrane request) to fetch chat transcripts and offline messages from HappyFox Chat—these are user-generated, third-party visitor contents the agent is expected to read and that could materially influence subsequent actions.