ClawHub
Back to skill

Security audit

subagent-testing

Security checks across malware telemetry and agentic risk

Overview

This skill is documentation for testing other skills with fresh conversations and does not add hidden execution, persistence, credential use, or destructive behavior.

Before installing, consider whether you want a testing-methodology skill to activate on generic testing or validation requests; otherwise its behavior is transparent and should mainly affect how the agent plans skill tests.

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
91% confidence
Finding
The trigger list contains very broad terms such as "testing," "validation," and "subagents," which are likely to match many normal user requests outside the narrow purpose of this skill. This can cause unintended auto-activation, increasing prompt-surface area and the chance that the skill influences conversations where it is not needed.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.