seedream-5-lite-image-edit

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow ingests outsider-authored free text directly from the user-provided --prompt argument and sends it to the AI Hive POST /openapi/v1/generation/image endpoint via cmd_image()/client.generate_image() (which also uploads user-supplied images), so prompt injection is possible.