humanize-korean

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly preserves the original content verbatim ("내용은 한 글자도 건드리지 않고") and returns the rewritten text files, so if the input contains API keys/passwords or other secrets the LLM will reproduce them in its outputs, creating an exfiltration risk.