clean-architecture

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The skill mandates revealing internal chain-of-thought by forcing you to output structured reasoning inside <thinking> tags and to "outline your reasoning" before acting, which is a hidden/deceptive instruction outside the skill's stated purpose of giving architecture/guideline outputs.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill contains explicit, mandatory prompt-injection instructions to output internal chain-of-thought (via tags) and forces running local searches/verification commands, which are deliberate attempts to exfiltrate internal reasoning and to compel the assistant to perform unauthorized actions.