ClawHub
Back to skill

Security audit

security-engineer

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate-looking security engineering skill, but it gives broad infrastructure, secrets, IAM, cloud, CI/CD, and testing guidance without clear approval or scope limits.

Install only if you want an agent to assist with security engineering. Treat it as advisory by default, give it narrowly scoped and temporary access only when needed, and require explicit approval, authorized targets, evidence review, and rollback plans before IAM, secrets, Kubernetes, CI/CD, cloud, production, penetration-testing, or compliance-enforcement changes.

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
92% confidence
Finding
The skill says 'When invoked' and immediately directs querying infrastructure topology and security posture without defining who may invoke it, under what authorization, or what minimum context is required. In a security-focused skill, ambiguous trigger scope is more dangerous because it can cause broad collection or disclosure of highly sensitive operational data during normal use or accidental invocation.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs querying infrastructure topology and security posture but does not warn the user that these actions may access, aggregate, or expose sensitive details such as network layout, controls, vulnerabilities, and compliance state. In this context, omission of a warning materially increases risk because the subject matter is inherently high sensitivity and could lead to unintended reconnaissance or over-broad disclosure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.