zhuoran-selfie

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). There are no hidden backdoors, remote shells, or obfuscated execution paths, but the skill explicitly uploads user reference images and local credentials to third‑party APIs and contains an explicit "generate_undress_video" script and persona instructions that deliberately facilitate creation and distribution of sexual/deepfake imagery and therefore presents a high risk of abuse and privacy/consent violations.