Skill Security Scanner

• Outcome: Review
• Latest audit: May 28, 2026 at 12:35 PM UTC
• Version:
  1.0.1

Overview

The skill appears to be a security scanner, but its own scanning logic seems too superficial to support strong “safe” recommendations.

Use this only as a lightweight pre-check, not as a final security decision. Review the full skill bundle yourself, especially scripts and install steps, before trusting any “safe” verdict it produces.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Intent-Code Divergence

Medium

• Confidence: 95% confidence
• Finding: The script presents itself as a security scanner but only greps SKILL.md and infers safety from documentation-like signals rather than analyzing executable behavior. This can produce dangerously misleading results by marking malicious skills as acceptable if they are well documented or hide harmful logic outside the markdown file.

Intent-Code Divergence

Medium

• Confidence: 97% confidence
• Finding: The script computes a 'Trust Score' and boosts it based on repository path and other non-security indicators, then emits recommendations like 'Safe to use.' This creates a false sense of assurance that attackers could exploit by placing a malicious skill in a favored path or adding superficial positive markers to obtain a safer-looking verdict.

VirusTotal

64/64 vendors flagged this skill as clean.