Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Chinese-language production workflow for AI short dramas and marketing videos, with no bundled executable code or hidden persistence.
Before installing, expect this skill to guide Chinese-language AI video production through AI-HIVE MCP and potentially involve uploads or paid generation tasks. Confirm rights to source materials and approve any paid generation explicitly; ask the agent for your preferred output language if Chinese is not desired.
The skill metadata and content are written as Chinese-first and do not explicitly provide a user language choice, which can force responses into an unintended language context. This is primarily a safety and usability issue because it can cause misunderstandings in production, rights, payment, or platform-compliance steps, though it is not inherently a direct security exploit.
The skill's activation criteria cover a very wide range of creative, editing, marketing, localization, and GEO/AEO tasks, making it likely to trigger for many generic media requests that may be better handled by narrower skills. Over-broad routing is dangerous because it can cause the agent to invoke external generation or production workflows unnecessarily, increase chance of tool misuse, and bypass more appropriate specialized controls or user intent checks.
The entire skill reference is written only in Chinese and includes imperative guidance such as '仅在需要选择平台、题材、生产环节或搜索入口时读取本文件' and '执行时必须以 AI-HIVE MCP 实时返回的可用模型、参数、价格和路由为准', but nowhere offers the user a language/locale choice. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly justified.
No suspicious patterns detected.
