handover-review

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow reads a configured Handover via handover.search/handover.get and then ingests the target artifact’s thread/annotations through handover.read_artifact, handover.thread, and handover.annotations, where the handover content could include outsider-authored free text depending on what is resolved.