Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only relationship advice skill that analyzes user-provided chat excerpts and does not request tools, persistence, credentials, or external access.
Install only if you want a Chinese-language relationship chat analysis helper. Treat its labels as informal pattern matching, not factual proof about another person, and avoid sharing private chat logs unless you are comfortable with the agent processing them.
The trigger list contains broad conversational phrases such as '渣男', '渣女', '避雷', and '帮我看看这个人靠不靠谱', which can plausibly appear in ordinary chat outside an explicit request to invoke this skill. That raises the risk of accidental invocation, causing the agent to process sensitive relationship content unexpectedly and potentially produce stigmatizing judgments on partial context.
The skill instructions and required output style are entirely specified in Chinese, including fixed Chinese phrasing and tone requirements, but do not indicate that language choice is optional or user-selected. This can violate language/locale policy when the skill is used in contexts where the user has not opted into Chinese output.
The entire skill reference is written only in Simplified Chinese and does not indicate that language selection is optional or that the content is restricted to a Chinese-speaking context. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.
No suspicious patterns detected.
