ClawHub
Back to skill

Security audit

仅适用于北京市企业职工基本养老保险退休金测算。支持依据北京政策参数计算基础养老金、个人账户养老金、过渡性养老金，并对未退休用户做未来缴费策略优化。

Security checks across malware telemetry and agentic risk

Overview

This is a local Beijing pension calculator that asks for sensitive retirement information and writes local result files, but its behavior matches its stated purpose and shows no hidden data transfer or destructive actions.

Install only if you are comfortable processing pension records locally with this skill. Provide only files you intend it to parse, avoid running it with elevated privileges, verify updated policy parameters against official sources, and delete generated tmp JSON/Markdown files after the calculation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill enables implicit invocation without any activation constraints, so the system may call this pension-calculation skill based only on loose semantic matching. That can cause the agent to solicit personal employment and retirement data unexpectedly or route unrelated conversations into this skill, creating privacy and integrity risks even though the domain is not inherently dangerous.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The script collects sensitive personal and quasi-financial data, including name, birth date, contribution history, unemployment history, and account balance, then writes it to a persistent JSON file under tmp without any warning, consent, retention policy, or access controls. On shared systems or developer workstations, this can expose personal data to other users, backups, logs, or later unintended reuse.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.