ClawHub
Back to skill

Security audit

代码安全审计员(免费版)

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed code-security review checklist with some broad wording, but no hidden persistence, exfiltration, or destructive behavior was found.

Install only if you want an agent to help review code and security configuration. Keep the review scoped to the intended project, avoid unnecessary exposure of secrets in .env files, and confirm before running commands that can change dependencies or project files, especially npm audit fix or npx npm-check-updates -u.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
78% confidence
Finding
Advertising generic create/query/export operations in a security-audit skill expands the apparent operational scope beyond read-only analysis and may prompt an agent to modify data or exfiltrate results without clear guardrails. In a skill with the exec tool enabled, ambiguous action verbs increase the chance of unsafe execution against user files, repositories, or exported artifacts.

Vague Triggers

Medium
Confidence
84% confidence
Finding
Overly broad trigger conditions can cause the skill to activate for routine requests that merely mention security, auditing, compliance, or encryption, leading to unintended use of read/exec capabilities. In an agent environment, over-triggering increases the risk of unnecessary file access, command execution, or disclosure of sensitive project context under the guise of a security review.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
Describing create/query/export behavior without warning that these actions may alter data or emit potentially sensitive audit results creates a safety gap. In a security-themed skill, users may assume analysis is passive, while an agent could interpret export or create operations as permission to write files, generate reports containing secrets, or interact with external systems.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.