english-polisher

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt requires echoing the user's original text verbatim in the report (the "原文" field), so if a user-provided input contains secrets (API keys, passwords, tokens) the model will reproduce them, creating an exfiltration risk even though the skill doesn't explicitly request credentials.