Skill Security Review

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:17 PM UTC
• Version:
  1.0.0

Overview

This is a transparent, instruction-only skill for reviewing other skills before installation, with one minor wording issue but no hidden code or unsafe behavior.

Reasonable to install as a review aid. Expect it to read and summarize skill artifacts before installation and to ask for confirmation before installing anything. Maintainers should replace the hard-coded user-specific wording with neutral language.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Natural-Language Policy Violations

Medium

• Confidence: 96% confidence
• Finding: The skill text explicitly centers decisions around a specific named user/context ("吴老板's machine and data") without any opt-in or parameterization. That can cause the agent to apply assumptions about identity, device ownership, language, or trust boundaries that do not generalize to other users, which may lead to incorrect security decisions or privacy leakage in multi-user/shared environments.

VirusTotal

65/65 vendors flagged this skill as clean.