ClawHub
Back to skill

Security audit

Humanizer AI

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only writing helper that rewrites text to sound less AI-generated, with no code execution, installs, credentials, network use, or persistence.

Install only if you are comfortable with a skill that can make AI-written text appear more human. Review rewrites for factual accuracy, voice, and provenance-sensitive contexts such as school, hiring, journalism, legal, or compliance work. Consider invoking it explicitly rather than allowing it to handle every generic polishing request.

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
88% confidence
Finding
The trigger list is broad enough to catch generic editing requests such as "polish text," which are not specific to AI-humanization. In an agent environment, this can cause the skill to activate outside its intended scope, overriding more appropriate writing or editing behaviors and potentially steering users toward evasion-oriented rewriting they did not explicitly request.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal