scanner-tuning

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s operate-phase workflow is explicitly concerned with analyzing vulnerability scanner results (including evidence, banners, plugin/check findings, and target-system responses) for false-positive classification and tuning, which means it ingests potentially outsider-authored free text coming from scan outputs/banners at runtime without needing first to select a specific item.