correlate

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill explicitly directs an automated agent to discover and document end-to-end attack chains—prioritizing account takeover, tenant escape, credential disclosure (e.g., "steal token") and other high-impact abuses—so while it contains no direct exfiltration or backdoor code, its design to produce detailed reproduction steps and to hunt for credential-theft/enabling chains creates a high potential for malicious misuse.