ClawHub
Back to skill

Security audit

Skill Evaluator

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed local evaluator for other skills and does not show hidden persistence, exfiltration, credential use, or destructive behavior.

Safe to install as a local skill review aid. Run it only on skill folders you intend to inspect, review any generated EVAL.md before publishing, and treat optional pip or npx tools as third-party dependencies you should install only in a trusted environment.

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 (1)

Vague Triggers

Medium
Confidence
84% confidence
Finding
The description contains broad trigger phrases like 'review, audit, evaluate, score, or assess a skill before publishing,' which overlap with many generic review requests. In agent routing contexts, overly broad triggers can cause this skill to be selected unexpectedly, increasing the chance that untrusted content is processed under an inappropriate rubric or that a more security-focused skill is bypassed.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.