finishing-development
Fail
Audited by Snyk on Jul 28, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (high risk: 1.00). The skill claims "仅验证，不做合并/清理" but later mandates automatically calling using-aidlc and auto-advancing to worker skills (e.g. MergeBack) without user confirmation, which is a hidden/deceptive instruction outside the stated scope.
Issues (1)
E004
CRITICALPrompt injection detected in skill instructions.
Audit Metadata