Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only product-design workflow skill with no evidence of code execution, credential access, exfiltration, persistence, or hidden behavior.
Safe to install for a Chinese-language internal product-design workflow. Before using its output as a real implementation plan, review the tracking examples for employee/payroll privacy, and fix or account for the packaged reference-path mismatch so all referenced steps load correctly.
The manifest description specifies activation on Chinese phrases such as "从0到1设计PRD/原型" and later requires user confirmation in Chinese ("确认"/"修改 X"). This imposes a specific language expectation without user opt-in or a documented justification that the skill is region- or locale-specific.
The entire skill file is written as mandatory operational guidance in Chinese, with no indication that other languages are supported or that the user can opt in to this locale. Under the policy, forcing a specific language without user choice is a natural-language policy violation unless the locale constraint is explicitly documented and justified.
The skill hard-codes Chinese-language prompts and examples throughout the step, requiring Chinese interaction without checking the user's language preference or offering an alternative. This can degrade usability, cause misunderstandings, and create exclusion/accessibility issues for users who are routed here but do not read Chinese, especially because the step explicitly requires collecting all four inputs before returning.
The file mandates Chinese-language output (for example, '输出 ✅/❌ 清单' and the entire validation spec is written as a required Chinese-format step) without indicating that the language is user-selectable or constrained by a documented business requirement. This can cause the agent to ignore a user's preferred language, reducing usability and potentially causing misunderstanding in a validation step, but it does not directly create a security exploit primitive.
The heading and required template content are entirely in Chinese, and L03 states the format must be followed strictly. This enforces a specific language/locale without any opt-in or explanation that the skill is region-specific, which matches the language-policy violation criteria.
The tracking plan explicitly includes sensitive payroll-related telemetry such as employee identifiers and salary release totals, but provides no privacy controls, minimization guidance, retention limits, or notice/consent considerations. In a company-internal HR/payroll system, these fields are highly sensitive and could enable employee monitoring, privacy violations, or secondary misuse if analytics data is broadly accessible or retained insecurely.
The file title and all instructional content are written in Chinese, and there is no indication that the user may choose another language or locale. Under the policy, forcing a specific language without opt-in is a natural-language policy violation unless clearly justified as region-specific.
The natural-language instructions force a single language presentation without indicating user choice or documenting a justified locale restriction. Under the policy rule, this can constitute a language/locale policy violation when no opt-in or scope limitation is provided.
No suspicious patterns detected.
