Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only short-video script writing skill with some usability caveats around broad triggers and language defaults, but no harmful system access.
Install this if you want help drafting short-video scripts and social platform copy, especially in Chinese. Specify the desired language, platform, audience, and tone to avoid unintended Chinese-only output or overly generic content-writing activation.
The skill advertises very broad trigger phrases such as '内容创作', '开场白', and '帮我写一个视频', which can overlap with ordinary conversation and unrelated writing requests. This increases the chance of accidental activation, causing the agent to invoke the wrong skill, override more appropriate instructions, or produce unintended content in mixed-context conversations.
The skill description and behavior are entirely Chinese-centric and do not state whether other user languages are unsupported or how language should be selected. In multilingual environments, this can lead to incorrect activation or unexpected Chinese-only outputs, which may confuse users and interfere with correct task routing, though it is primarily a reliability and usability issue rather than a direct security exploit.
This markdown file forces a specific language/locale for all users through its natural-language content, and there is no indication that users can choose another language or that the Chinese-only scope is justified. The policy for this audit explicitly calls out language or locale constraints without user opt-in as a violation.
Line L124 states '标题要求（英文）', which imposes English-language output for YouTube content. The file does not provide an opt-in, alternative language guidance, or a documented reason that the skill must constrain users to English, which conflicts with the policy against forcing a specific language without user choice.
This markdown file is entirely written as Chinese-only script templates and presents the skill content as if that language is required by default. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation when no alternative language choice or justification is provided.
No suspicious patterns detected.
