ClawHub
Back to skill

Security audit

Epistemic Hygiene

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only reasoning skill that encourages more careful research and discussion habits, with no code execution or hidden data access found.

Install only if you want the agent to automatically apply this discussion style during strategy, product, and research conversations. Be aware that its verification guidance may cause the agent to search the web using terms from your prompt, so avoid including confidential project names or sensitive details in queries unless your host agent redacts them.

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 for short replies matches extremely common responses like "ok", "yes", and "go", causing the skill to auto-activate on minimal acknowledgments. That can make the agent continue with strong assumptions instead of recalibrating intent, which is especially risky in a skill that governs reasoning style and conversational framing.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.