product-inspection

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

• Third-party content exposure detected (medium risk: 0.30). In the required runtime workflow, the agent explicitly reads outsider-authored free text such as user-provided product inputs/exports and potentially support/feedback content via the “fallback chain” evidence sources (exports & screenshots, interview, and “Support & feedback” assets) during the stations’ live/code/export/review procedures.