ClawHub
Back to skill

Security audit

Feynman Technique Explainer

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only teaching skill that formats explanations and quizzes, with no evidence of code execution, credential access, persistence, or data exfiltration.

Reasonable to install as a teaching aid. Expect it to simplify concepts and ask quiz questions; consider narrowing its triggers if your agent has specialized lookup, medical, legal, financial, or technical-answer skills, and do not rely on it alone for safety-critical guidance.

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
91% confidence
Finding
The listed activation phrases are very generic and overlap with normal user requests such as 'Explain X to me' and 'Help me understand X.' In an agent with multiple skills, this can cause the skill to activate too broadly, intercept unrelated conversations, and override more appropriate or safer domain-specific handling, especially because it claims to work on 'any subject' including medicine and technology.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger phrases are very broad (for example, common requests like 'what is X?' or 'break this down for me'), so this skill may activate for many ordinary queries where a simpler or more appropriate skill should respond. Over-triggering can cause incorrect routing, unnecessary constraint application, and degraded agent behavior, especially because this skill imposes strict formatting and content rules that may not fit the user's intent.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.