Security audit
Security checks for vulnerabilities and agentic risk
This is a Markdown-only Chinese sci-fi survival roleplay skill with no code execution or data access, but users should treat its survival scenes as fiction rather than real-world guidance.
Install only if you want a primarily Chinese sci-fi roleplay experience. Do not treat its alien food, water purification, medical, or survival content as real-world instructions; it should be used as fiction. There is no indication that it runs code, accesses credentials, or persists changes.
This markdown file presents all user-facing instructions and descriptions in Chinese, and does not indicate that the user can choose another language. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is clearly documented and justified.
The file is written in a fixed mixed Chinese-English style and directs the roleplay entirely in that language pattern, but it does not indicate that this is optional or limited to a particular audience. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy concern.
The markdown explicitly defines trigger conditions, but several options such as '在探索中偶遇' and '收到神秘信号并追踪' are very broad and could match many ordinary story situations. The file does not provide exclusion conditions or clear constraints for when this skill should activate versus when a different chapter should be used.
The description and the rest of the skill are written primarily in Chinese, and the file does not indicate that language choice is optional or that the skill is intended only for a Chinese-speaking or region-specific context. Under the policy, a fixed language/locale without user opt-in can be a natural-language policy violation.
The content presents self-experimentation with unknown alien fruit as an interactive option, including skin testing and taking a small bite, before any strong safety framing. Even though the narrative later shows poisoning and says not to do it, the scene still models hazardous trial-and-error behavior that a user could imitate or normalize, especially because the protagonist is framed as scientifically competent.
The file’s description and nearly all instructional content are written in Chinese or mixed Chinese-English, which effectively constrains the skill to a specific language. There is no indication that users may choose another language or that the locale restriction is required for a region-specific purpose.
This markdown skill is written in a hybrid Chinese-English style across the description and dialogue, effectively imposing a specific mixed-language output pattern on users. The file does not offer a language preference option or explain that the locale is intentionally constrained, which fits the language/locale policy violation criteria.
The chapter frames consumption of unknown alien plants, fruits, and animal protein as playful experimentation without an explicit safety boundary. Even in fictional roleplay, this can normalize hazardous ingestion behavior and encourage users to treat uncertain edibility as acceptable, especially because the text repeatedly presents curiosity and hunger as sufficient justification.
The stew scene depicts tasting a mixture built from uncertain ingredients as a normal trial-and-error cooking step. That presentation is risky because it models sampling unknown food despite no safety confirmation, reinforcing unsafe experimentation rather than caution.
The salad scene treats tongue tingling after eating as a quirky positive feature rather than a possible toxicity or allergic reaction. This is dangerous because it reframes a classic warning sign of adverse exposure as success, which could desensitize users to symptoms that should trigger stopping consumption and seeking help.
The meat preparation sequence presents trapping, butchering, cooking, and eating an alien animal as safe once it 'looks good' and a scanner shows internal temperature, but it omits broader contamination and biological hazard concerns. That can mislead users into thinking cooking appearance or a single temperature check is enough when unknown pathogens, toxins, or cross-species hazards may remain.
The skill content is predominantly written in Chinese and implicitly steers the interaction into Chinese without offering the user a language choice. This can exclude or mislead users who expect responses in another language, reducing informed consent and creating accessibility and usability risks, though it is not a direct code-execution or data-exfiltration issue.
The skill hard-codes a Chinese-speaking persona and mixed Chinese-English response style without indicating that the assistant should adapt to the user's preferred language. This can override user expectations, reduce accessibility, and cause unsafe or unusable interactions in contexts where the user cannot understand the output, though it is not a direct security exploit.
The skill content is written primarily in Chinese and implicitly steers the interaction into that locale without an explicit user preference check. While not a classic security flaw, it can override user expectations, reduce transparency, and make it easier to hide behavior or important instructions from users who do not understand the forced language.
The skill explicitly directs the agent to speak mostly Chinese ('说话尽量用中文') without checking or adapting to the user's language preference. This can degrade accessibility and cause exclusion or misunderstanding for users who do not read Chinese, especially during urgent or emotionally intense interactions where comprehension matters most.
The file’s description and all user-facing content are written primarily in Chinese, and the skill provides no indication that users may choose another language or that the Chinese-only presentation is required for a specific regional purpose. Under the language/locale policy, a skill should not implicitly force a specific language without user opt-in or clear justification.
The description is written as a Chinese-language skill title and framing, and the file consistently presents the interaction in a fixed Chinese/mixed-language style without indicating that the user can choose another language. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy violation.
The skill metadata and content are written primarily in Chinese and do not provide a mechanism to adapt to the user's preferred language. This can cause misunderstanding of safety-critical or task-relevant instructions, reduce accessibility, and create exclusion or misuse risk for users who cannot read the language well. In this game context the impact is limited because the content is narrative rather than operationally dangerous, but it is still a genuine quality and safety issue.
The skill gives concrete survival guidance about testing, boiling, filtering, chemically treating, and then drinking unknown water, including confidence-building language like 'Safe to drink!' without any explicit disclaimer that this is fictional and unsafe to copy in real life. Because users may generalize these steps to real emergency situations, the content could encourage hazardous behavior when dealing with contaminated water, especially since the guidance oversimplifies what purification can and cannot guarantee.
The description explicitly defines the skill in Chinese and the body consistently instructs a code-switched Chinese-English style, indicating a fixed language/locale behavior. Under the stated policy, forcing a language style without user opt-in is a natural-language policy concern unless the locale constraint is clearly justified.
The description and all narrative instructions are written in Chinese, with no indication that users may choose another language or that the skill is intentionally limited to Chinese-speaking users. Under the language/locale policy, a forced language without opt-in can be a natural-language policy violation.
No suspicious patterns detected.
