Security audit
Security checks for vulnerabilities and agentic risk
This skill performs disclosed retail inventory analysis and report generation, with no evidence of hidden execution, exfiltration, destructive actions, or persistence beyond expected report files.
Install this only where the metric-query dependency is trusted and authorized for the relevant inventory and sales data. Expect it to create HTML reports and an XLSX action plan that may contain commercially sensitive inventory, pricing, brand, and season data, so store and share those files according to your business data rules.
The trigger conditions are extremely broad and include common phrases like asking whether inventory has problems or what products should be discounted. This can cause the skill to activate unexpectedly in conversations where the user did not clearly intend to run a data-analysis workflow, increasing the chance of unnecessary data access, external queries, and file generation.
The skill is written entirely in Chinese and specifies trigger phrases only in Chinese, implying operation in a single language without opt-in or alternative locale support. Under the policy, a forced language/locale is a violation unless the skill offers a user choice or clearly documents a justified region-specific constraint.
The skill explicitly depends on an external query capability to retrieve business metrics, but does not warn the user that inventory and sales-related data may be transmitted to another component or service. This matters because retail operational data can be commercially sensitive, and silent transfer to external dependencies weakens user awareness and informed consent.
The skill requires generating HTML outputs, and later XLSX files, without first informing the user that files will be created or explaining where they will be stored and who can access them. In a business context, these files may contain sensitive inventory, pricing, and brand-performance data, creating persistence and exposure risks beyond the immediate chat session.
The module docstring states the code is for direct copying/execution and all user-facing natural-language text is written exclusively in Chinese. This creates a language/locale constraint without any opt-in or indication that the skill is intended only for a Chinese-language context, which matches the policy's language-choice concern.
The entire skill reference is written in Chinese and does not indicate that language selection is optional or user-configurable. Under the policy, forcing a specific language without user opt-in is a natural-language locale violation unless the regional constraint is explicitly documented and justified.
The document declares lang="zh-CN", and all visible interface text in the template is written in Chinese, which indicates a fixed language/locale choice. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the regional constraint is clearly documented and justified, which is not present in this file.
The title and content specify the report format entirely in Chinese and present it as the required final HTML report specification. This creates a language constraint without any opt-in, alternative locale, or documented justification for why the skill must operate only in Chinese.
No suspicious patterns detected.
