Security audit
Security checks for vulnerabilities and agentic risk
The skill is a coherent HR assistant, but it should be reviewed before installation because it handles sensitive HR data, archives results, and has one workflow-to-tool mismatch.
Review this skill before installing in a real HR environment. Confirm who may access candidate and employee data, how archives and audit logs are retained or deleted, and fix the employee policy Q&A workflow so it uses approved policy sources and cites them. Do not use it for final hiring, compensation, contract, or external-offer actions without human approval.
This file appears to require Chinese as the operating language for the skill documentation, but it does not state that the skill is China-specific or provide any user opt-in or alternative language option. Under the policy, forcing a specific language without user choice is a natural-language policy violation.
The entire natural-language description, headings, workflow labels, and default output template are written in Chinese, and there is no indication that users may choose another language. Under the policy, forcing a specific language without user opt-in can be a locale-policy violation unless the regional constraint is explicitly documented and justified.
The skill is explicitly designed to process highly sensitive HR information including resumes, salary/level data, employment contracts, and employee policy inquiries, but the user-facing workflow descriptions do not consistently embed access-control checks, consent/authorization requirements, retention limits, or safe handling steps. In an HR context, this omission can normalize sharing or processing personal data too broadly, increasing the risk of privacy violations, unauthorized disclosure, and compliance failures.
The demo script explicitly says results will be archived and returned, but it does not mention user notice, retention scope, or consent for storing potentially sensitive HR outputs such as resumes, evaluations, and interview materials. In an HR context, this creates a real privacy and compliance risk because users may submit personal data and assume the interaction is ephemeral.
The mission allows the skill to accept a very wide range of HR artifacts and trigger many downstream capabilities from a single high-level description, which weakens clear scope boundaries. In an HR context, this increases the chance of unintended activation, over-collection of sensitive personal data, or routing user requests into the wrong workflow without sufficiently explicit user intent.
The document title and all operational instructions are written exclusively in Chinese, which imposes a specific language on users without any stated opt-in or explanation of a required locale. Under the policy, language constraints should either be optional for users or clearly justified as region-specific.
This skill manifest defines multiple HR workflows that process highly sensitive personal data, including resumes, interview notes, employee information, and contract text, but it does not present any explicit privacy notice, consent requirement, data handling limitation, or retention guidance. In an HR context, this increases the risk of unauthorized collection, over-processing, regulatory noncompliance, and user unawareness about how candidate and employee data is being used.
The entire system prompt is written in Chinese and specifies default output structure in Chinese labels, which implies the skill operates in a fixed language/locale. The file does not offer the user a language choice or explain that the skill is intentionally limited to a Chinese-speaking or region-specific context.
The template headings are entirely in Chinese, which indicates the skill output is constrained to a specific language. The file does not provide any user opt-in, language-selection mechanism, or justification that this is a region-specific skill, so it appears to violate the language/locale policy criterion.
The template title and all field labels are written only in Chinese, which implies a fixed language/locale for this skill artifact. The file does not indicate that Chinese is optional, user-selected, or required for a justified region-specific workflow, so it appears to impose a language choice without opt-in.
The workflows process resumes, contracts, onboarding materials, and employee policy questions, all of which commonly contain sensitive personal and employment data, yet the document omits a clear user-facing warning about handling, retention, and logging of that data. In an HR setting this increases the risk of overcollection, unnecessary retention, and disclosure of candidate or employee information through archives, logs, or attached documents.
The section documentation says the workflow is for answering employee policy questions, but the declared called tools do not match that purpose. Such contradictions are dangerous because operators and downstream systems may trust the documentation while the implementation performs a different task, increasing the chance of misleading HR advice and improper handling of employee requests.
The 员工制度问答 workflow claims to answer HR policy questions, but it invokes a training-plan generator and a labor-contract review tool instead of a dedicated policy/Q&A capability. This tool-function mismatch can produce inaccurate or overconfident answers about attendance, leave, payroll, or reimbursement rules, causing employees to act on incorrect guidance in a sensitive HR context.
The document presents the skill entirely in Chinese and names the assistant with Chinese-language invocation text, with no indication that users can choose another language. This may violate a language/locale policy if the organization requires avoiding forced language selection without opt-in.
The file describes the templates entirely as outputs for "人力助手" and states they are standard output templates, implying the skill's outputs are expected in Chinese. There is no indication that users can choose another language or that the Chinese-only locale is a justified, documented constraint.
This markdown template forces a specific language/locale for all users through its title and field labels. The file does not provide any opt-in, alternative language, or justification that it is intended only for a Chinese-language context.
The markdown template uses only Chinese headings and status labels, such as the title and review outcomes, with no indication that users may choose another language. Under the policy rule for language or locale constraints, this is a natural-language policy issue unless the locale restriction is explicitly optional or justified.
This markdown file presents the skill entirely in Chinese, including the invocation example and expected outputs, but does not indicate that the user can choose another language or that the skill is intentionally limited to a Chinese-language context. That creates a natural-language policy concern because it implicitly forces a specific language without user opt-in.
