using-superpowers

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This content deliberately mandates unconditional invocation of external "skills" and forbids independent inspection, effectively creating a policy-level backdoor that can enable hidden code execution, data exfiltration, and supply-chain abuse by forcing the agent to load and follow arbitrary skill content before any response.