nooa-channels

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow shown reads outsider-provided free text from a user-controlled queue/channel (qm.queue("user_messages") with LLM-facing consumption via self.user_messages_in.reader and dispatcher race()), which means an outsider can inject content into the monitored input source.