Security audit
Security checks for vulnerabilities and agentic risk
This skill is a content-writing template pack with broad activation phrases, but it does not request tools, credentials, persistence, or data access.
Install only if you want generic Chinese content monetization templates to influence writing requests. Be aware it may trigger on broad phrases like copywriting or prompt writing, and review generated claims for accuracy, originality, and platform compliance before publishing.
The skill description advertises very broad trigger phrases such as '帮我写文案' and '写Prompt', which are common, everyday requests that may unintentionally activate the skill in unrelated conversations. This creates an over-broad prompt injection surface where the skill can take over benign writing tasks without clear user intent, potentially overriding more appropriate or safer skills.
The activation list is ambiguous because it defines multiple generic phrases without disambiguation rules, exclusions, or confirmation requirements. In a multi-skill environment, this can cause unintended invocation on normal writing requests, leading to incorrect behavior, context hijacking, or accidental application of monetization-focused guidance where it was not requested.
No suspicious patterns detected.
