switch

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the assistant to export "every memory" and "preserve my words verbatim" (including personal details and stored context), which would force the model to output any stored secrets or API keys verbatim if they exist, creating direct exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill's export prompt explicitly instructs a full, verbatim dump of all stored memories, context, preferences, and instructions (including personal details and likely secrets), which creates a high-risk data-exfiltration/credential-disclosure capability even though there is no code-level backdoor, RCE, obfuscation, or network exfiltration mechanism embedded.