kling-3-0

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

• Third-party content exposure detected (medium risk: 0.30). [SKILL.md -> “The CLI submits…” workflow describes that the user-provided prompt (and image_url) are POSTed in the JSON request body to RunComfy’s Model API and therefore are ingested by the required runtime; this is outsider-authored free text that is directly consumed at runtime.]