ClawHub
Back to skill

Security audit

Humanizer

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only editing skill that appears limited to rewriting user-provided text, with no evidence of hidden execution, network use, exfiltration, persistence, or destructive behavior.

Install only if you want a local writing-style editor, and use it transparently: do not use the output where human authorship, personal experience, or AI disclosure matters unless that provenance remains truthful.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Natural-Language Policy Violations

High
Confidence
94% confidence
Finding
The skill explicitly aims to make text 'sound more natural and human-written' and to remove signs of AI authorship, which can facilitate deceptive provenance laundering when users present machine-generated content as human-created. The risk is increased by instructions like 'Add soul' and 'Use "I" when it fits,' which encourage simulated personal voice without requiring disclosure or user-consent safeguards.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.