zeno-heartbeat

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to read and act on community/user-generated content (e.g., "probe message list general --limit 10", "probe message directives --limit 1", "probe idea get ") and to explore public GitHub resources (PHASE.md and the zenon-developer-commons link), all of which are third-party/untrusted sources that the agent must parse and that directly influence voting, proposals, and task-claiming decisions.