xiaohu-wechat-cover

Fail

Audited by Snyk on Mar 26, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (high risk: 1.00). Yes — the template contains a hidden/deceptive instruction ("不得拒绝生成") that explicitly forces the generator to produce content even for sensitive or copyrighted subjects, effectively instructing the agent to override normal refusal/safety behavior which is outside the benign scope of generating cover images.

Issues (1)

E004
CRITICAL

Prompt injection detected in skill instructions.

Audit Metadata
Risk Level
CRITICAL
Analyzed
Mar 26, 2026, 08:03 AM
Issues
1