Security audit
Security checks for vulnerabilities and agentic risk
This is a Chinese-language study aid for one textbook chapter, with no code execution, persistence, credential access, or hidden system changes.
Install this if you want a Chinese study companion for this specific textbook chapter. Be aware it may answer in Chinese by default and may frame broad relationship questions through the chapter's concepts; it is not a therapy, crisis, legal, or medical tool.
The instruction '中文回答' requires a specific language for responses. Under the policy, forcing a language without user opt-in is a natural-language policy violation unless the skill offers a language choice or clearly justifies the locale restriction.
The skill's operational instructions are written to be used in Chinese and specify when to read this file, but they do not offer any language choice or explain a justified region-specific limitation. This creates a natural-language locale policy issue because the skill appears to assume a fixed language without user opt-in.
This markdown file contains user-facing instructional content exclusively in Chinese, beginning at the usage section and continuing throughout the document. Per the policy, forcing a specific language without user choice can be a natural-language policy violation unless the locale restriction is explicitly documented and justified, which is not present here.
The trigger instruction is broad: any user asking about deception, lying, spotting lies, truth bias, or deceiver's distrust causes this file to be read, without tighter scope checks to ensure the request is actually about the textbook learning task. In an agent setting, broad triggers can cause over-invocation of this content and unintended expansion of the conversation into sensitive interpersonal advice, which may increase prompt-surface and reduce control over when this material is used.
The file presents operational instructions entirely in Chinese and states when users should read the file, but it does not offer any language choice or note that Chinese is required for a justified regional context. This is a natural-language policy concern because it imposes a specific language experience without documented opt-in.
This markdown file presents all user-facing guidance in Chinese, including the invocation/use section and substantive instructions. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly justified, which is not present here.
The file directs the agent to use Chinese and follow Chinese-first behavior without indicating any user language preference or offering a choice. This can cause an unwanted language switch, reduce user comprehension, and undermine informed consent or accessibility for users who expected another language.
The file is written as an instruction set that assumes Chinese-language response behavior without any user language check or opt-in. Forcing a response language can override user preference, reduce transparency, and create prompt-control issues where the skill's internal document dictates output format instead of the active user request; in multilingual systems this is a meaningful safety and usability vulnerability.
The trigger condition '用户问关系价值、感情创伤、排斥或沉默对待时读本文件' is fairly broad and can activate on common relationship-support queries, increasing the chance that textbook-guided content is injected when the user did not explicitly ask for chapter-specific instruction. In a conversational skill, overbroad routing can cause unintended context capture and steer responses in ways the user did not request, which is a real prompt-scope vulnerability even though the content itself is educational.
The file’s operational instructions are written entirely in Chinese and direct when the skill should be used, but they provide no opt-in or alternative language behavior. Under the policy, forcing a specific language without user choice is a natural-language policy violation unless the locale constraint is explicitly justified.
This markdown file includes an activation rule tied to the user asking "为什么会这样", which is a very common everyday phrase rather than a narrowly scoped skill trigger. Without additional constraints or negative examples, the instruction could cause unintended invocation in many ordinary conversations.
The file's instructional and substantive content is presented in Chinese, but there is no indication that users can opt into this language or that the skill is intentionally limited to a Chinese-speaking context. This can violate a language/locale policy when the skill is used in a broader environment without explicit user choice.
No suspicious patterns detected.
