ClawHub

Ai Humanizer

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed text-rewriting skill with no executable code or system access, though its “undetectable” framing creates misuse risk.

Install only if you want a style-editing helper and can control when it is used. Do not use it to hide AI authorship where disclosure is required, impersonate someone, evade platform rules, or alter text where exact wording matters.

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 (3)

Vague Triggers

Medium
Confidence
84% confidence
Finding
The README states the skill activates automatically when it detects 'rewriting/humanizing intent,' but does not define narrow trigger boundaries or exclusions. In an agent environment, this can cause the skill to engage on broad editing requests and silently alter content style or meaning in contexts where such rewriting is inappropriate, including compliance, legal, or safety-sensitive text.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill activates on very broad prompts such as requests to 'humanize,' rewrite, or make text 'undetectable,' which can capture many ordinary writing tasks without clear user consent boundaries. In this context, the ambiguity is more concerning because the skill is explicitly designed to make AI-generated text harder to detect, increasing the chance of misuse for evasion, impersonation, or academic/workplace policy bypass.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The description openly promotes making content 'undetectable' without any warning, limitation, or acceptable-use language. That omission materially increases risk because users are guided toward concealment of AI authorship rather than ordinary copyediting, which can facilitate deception in academic, professional, compliance, or platform-moderation contexts.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal