Back to skill

Security audit

企业人事技能包

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent HR reference skill, but users should verify its payroll and attendance formulas before using them for real employment decisions.

Install only if you want a Chinese/PRC-oriented HR reference pack. Treat its legal tables, payroll calculations, and Excel/VBA snippets as templates, not authoritative outputs; verify local law, company policy, and the attendance formulas before using them with employee data.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill description and most triggers are written in Chinese, presenting the skill as operating in a specific language/locale, but the file does not indicate that this is optional or user-selectable. Under the policy, locale or language restrictions should either be user opt-in or clearly documented as justified for a region-specific tool.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list includes many generic HR terms such as 'HR', '人事', '工资', and '招聘', making the skill likely to activate for ordinary conversations that may not need this specific skill. Over-broad activation is a security and safety issue because it increases unnecessary exposure to the skill's instructions, formulas, and automation snippets in unrelated contexts.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
99% confidence
Finding

The attendance formula comment labels COUNTIF(...,"<9:00") as 'late count', but that condition actually counts punches earlier than 9:00, not later. In an HR/payroll skill, this can directly distort attendance, disciplinary actions, and wage deductions, creating compliance and employee-relations risk if copied into real spreadsheets.

Content

No source excerpt is available for this finding.