Back to skill

Security audit

社保公积金基数与比例查询

Security checks for vulnerabilities and agentic risk

Overview

The skill looks like a local prototype but presents itself as a current nationwide payroll-policy tool, which could mislead users relying on it for benefits or compliance decisions.

Treat this as a demo or prototype, not an authoritative payroll or compliance tool. Before installing for real use, require the publisher to limit the description to the actual covered cities or implement verified official-source retrieval with source URLs, last-updated timestamps, cache age, and clear refresh controls.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The skill claims nationwide, up-to-date policy coverage with automated updates, but the described implementation reportedly relies on limited hardcoded data and lacks a real authoritative update pipeline. This is dangerous because users may rely on inaccurate or stale social security and housing fund figures for payroll, compliance, or benefits decisions, causing financial and regulatory harm while the skill presents itself as current and comprehensive.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill claims to provide latest nationwide social insurance and housing fund data with automatic updates, but the implementation only returns hard-coded mock data for three cities and does not actually fetch current official data. In a payroll, HR, or compliance context, users may rely on inaccurate or incomplete contribution bases and rates, causing legal, financial, and operational harm.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill advertises behaviors that require file read/write and network access, but it does not declare any explicit tool scope or permissions boundary. This is dangerous because users and platform controls cannot clearly understand or constrain what the skill may access, increasing the risk of unintended local file modification, data persistence, or outbound requests.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The code comments and behavior imply that saving the cache keeps data 'always latest', but the cache only persists static mock content. This can mislead users and downstream systems into trusting stale or fabricated data as freshly updated, compounding the risk of compliance errors.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill caches query results locally for up to 30 days but does not clearly warn users in the description that outputs may come from local cached data rather than fresh retrieval. This is dangerous in context because users may assume the results are current official figures, leading to stale policy data being used for employment, payroll, or benefits decisions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This code contains user-facing help text, status messages, and comments entirely in Chinese, including the CLI description and argument help. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly justified.

Content

No source excerpt is available for this finding.