Security audit
Security checks for vulnerabilities and agentic risk
This is a prompt-writing skill for GPT-image2 with broad but disclosed routing behavior and no evidence of code execution, credential access, persistence, or data exfiltration.
Installers should know this skill may take over fairly general image-prompt requests and will usually produce both Chinese and English prompt text for GPT-image2. It does not appear to request sensitive access or modify the system.
The trigger conditions are broad enough to match generic image-generation requests such as '生成一张海报' or '写个提示词', which can cause this skill to activate when the user did not explicitly ask for this specific framework. In a multi-skill environment, that can lead to unintended routing, overriding more appropriate skills, and surprising the user with constrained output behavior tied to GPT-image2 only.
The file states that output should include both Chinese and English and that the English version is used for actual GPT-image2 input, making a language choice on the user's behalf. This is a natural-language policy concern because the skill does not clearly frame English output as optional or user-selected.
This markdown template library is predominantly written in Chinese, but several templates switch to imperative English prompt content such as 'Create a premium...' and 'Premium product shot...' without indicating that users may choose their preferred language. That can conflict with a language/locale policy requiring user opt-in before forcing a specific language.
No suspicious patterns detected.
