Security audit
Security checks for vulnerabilities and agentic risk
This skill drafts Chinese merchant review responses from user-provided reviews and clearly requires human review before anything is posted.
Install only if you want Chinese-language draft handling for merchant reviews. Review every generated response before posting, especially high-risk food safety, injury, legal, or counterfeit complaints, and avoid sending customer personal information unless it is necessary and can be masked.
The skill description is entirely in Chinese and presents example invocation text only in Chinese, which signals that use is expected in a specific language. The file does not mention any language choice, opt-in, or region-specific limitation that would justify the constraint.
The description states '本技能面向中文评价设计', which imposes a language constraint as part of the skill definition. The file does not offer a user language choice or explicit opt-in for Chinese-only handling, so this is a natural-language locale policy issue under the stated rules.
The markdown content, including the title, guidance, and sample output, is entirely in Chinese and does not indicate that the user can choose another language or that the skill is intentionally limited to a Chinese-speaking context. Under the stated policy, forcing a specific language without opt-in is a natural-language policy violation.
The entire template prescribes response style and wording in Chinese, including specific Chinese phrases to use or avoid, but does not state that the skill is limited to Chinese-language merchants or users. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.
The entire template is written as a prescriptive response format for e-commerce review handling in Chinese, with no indication that users may choose another language or locale. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy violation unless the locale limitation is explicitly documented and justified.
No suspicious patterns detected.
