source-verification

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly describes and scripts interacting with public, untrusted sources—e.g., social media account analysis, reverse-image searches (Google Images, TinEye, Yandex) and the archive_url example that calls https://web.archive.org/save/{url}—so it would ingest and interpret arbitrary third-party content as part of its workflow.