Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent Chinese-language Yorkie care assistant that asks for pet-care details and uses web search for current information, with no evidence of hidden execution, persistence, exfiltration, or unsafe authority.
Install only if you want a Chinese-language pet-care assistant. Share only pet and household details you are comfortable providing, expect search queries to be sent to the configured web-search provider, and use a veterinarian for urgent or serious health symptoms.
The manifest sets language: zh-CN, and the skill description is entirely presented in Chinese without stating that users may choose another language. The policy requires flagging language or locale constraints when a specific language is imposed without explicit user opt-in or clear region-specific justification.
The instructions explicitly require prioritizing Chinese search keywords and prescribe a Chinese example closing line for every reply, which imposes a language/locale preference without user opt-in. Under the policy, language constraints should be optional, user-selected, or clearly justified as region-specific.
The manifest explicitly sets the skill language to "zh-CN", which indicates the skill is intended to operate in a specific language/locale. There is no accompanying text offering a language choice or documenting a justified region-specific constraint, so this appears to violate the language/locale policy requirement.
This markdown skill description instructs the assistant to gather structured information such as feeding schedule, medical history, deworming, and vaccination timing immediately at first contact. While not highly sensitive like credentials, it is still user-associated household and animal health information, and the description provides no disclosure or caution about requesting, retaining, or using that data.
No suspicious patterns detected.
