Security audit
Security checks for vulnerabilities and agentic risk
This is a Chinese-language AI course tutor skill with broad teaching triggers, but it contains only markdown guidance and no hidden execution, persistence, or data exfiltration behavior.
Install this if you want a structured Chinese AI-learning tutor. Expect it to ask assessment questions and push a fixed course workflow when AI-learning topics come up; users who only want brief Q&A may find the trigger scope too aggressive.
The skill metadata states it should auto-activate whenever a user expresses broad interest in learning AI, which is an overly permissive trigger. In practice this can hijack ordinary conversations, causing the agent to switch into a rigid workflow unexpectedly and potentially collect unnecessary user profiling information before the user explicitly asked for the skill.
The trigger section explicitly says the full standardized teaching flow should start automatically without extra confirmation for a wide range of common phrases. Because these phrases are ambiguous and common in normal Q&A, the skill can activate without informed user intent, leading to unwanted behavior, excessive questioning, and reduced user control over agent actions.
The listed technical phrases include many generic topics such as 'machine learning,' 'Transformer,' 'RAG,' and 'Prompt Engineering,' which are common discussion subjects and not reliable indicators that the user wants this skill engaged. This makes accidental activation more likely and is more dangerous here because the skill imposes a mandatory, non-skippable flow that can override the conversational context and push the user into a predefined process.
The entire skill content is written exclusively in Chinese and presents all guidance, labels, and teaching instructions in that language. There is no indication that the user can opt into another language or that the skill is intentionally limited to a Chinese-speaking or region-specific audience, which creates a natural-language locale policy concern.
The title and the entire instructional content are written exclusively in Chinese, with no indication that other languages are supported or that the language choice is optional. Under the policy, forcing a specific language without user opt-in is a natural-language locale violation unless clearly justified as region-specific.
Suspicious Unicode normalization or mixed-script content
The file is written entirely in Chinese and presents itself as a general teaching/reference document, but it does not indicate that Chinese is optional, user-selected, or required for a region-specific purpose. Under the policy rule for natural-language violations, this can be interpreted as forcing a specific language without user opt-in.
The file title and directive are written as a strict standard template in Chinese, and L003 says every chapter must strictly follow this structure. That creates a natural-language locale constraint without user opt-in or an explicit justification that this skill is intended only for a Chinese-language context.
This markdown file is entirely presented in Chinese and the title explicitly frames it as the authoritative course outline, without any indication that other languages are available. Under the policy rule for language or locale constraints, forcing a specific language without user opt-in is a natural-language policy concern.
