nooa-middleware-hooks

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

• Third-party content exposure detected (low risk: 0.10). The runtime workflow described in SKILL.md uses middleware hooks that can access the rendered LLM prompt (agent.event_manager.intercept("llm_call", ...) wraps runtime.generate() and exposes ctx.messages), but the document does not show any outsider-authored free-text ingestion path where user-submitted content is read without first selecting a trusted item.