eachlabs-face-swap

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

• Third-party content exposure detected (high risk: 0.80). The skill's prediction inputs accept arbitrary public image/video URLs (e.g., the "target_image", "swap_image", and "target_video" fields shown in the curl examples) which the EachLabs API fetches from the open web, meaning the agent ingests untrusted third-party content as part of its workflow.