using-superpowers

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This skill content is a high-risk prompt-injection/jailbreak that maliciously instructs the agent to always invoke external "skills", prohibits normal inspection (e.g., "Never use the Read tool"), and thus can be used as a covert backdoor to load and execute arbitrary malicious skill payloads (enabling data exfiltration, credential capture, or remote code execution) even though it contains no direct network/execution code itself.