Security audit
Security checks for vulnerabilities and agentic risk
This is a straightforward Japanese-to-Chinese translation skill with minor activation-scope issues but no evidence of hidden, persistent, destructive, or data-exfiltrating behavior.
Reasonable to install for Japanese-to-Chinese business or legal translation. Be aware that its broad triggers may activate for generic translation requests, so confirm the target language when working with sensitive documents.
The trigger list is broad and ambiguous, including generic terms like '翻译', '日语文件', and business-document keywords such as 'LOI'. This can cause the skill to activate in contexts the user did not intend, leading to incorrect routing of sensitive content, unwanted disclosure to the skill, or interference with safer/more appropriate skills.
The skill hardcodes Chinese output without checking whether the user wants another target language or a bilingual result. In a translation workflow, this can cause usability failures, misunderstanding, or accidental disclosure if translated material is shared with recipients expecting a different language, though it is not a direct security exploit by itself.
No suspicious patterns detected.
