ClawHub
Back to skill

Security audit

Humanizer Zh

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Chinese text-editing skill for rewriting AI-like prose, with no hidden code, credential access, or background behavior found.

Use this only on documents you intend to have read and rewritten. Avoid applying it to legal, academic, compliance, or disclosure-sensitive text unless you confirm that citations, provenance statements, and required AI-use disclosures remain intact.

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
87% confidence
Finding
The trigger condition "编辑或审阅文本，去除 AI 写作痕迹" is broad enough that the skill may be invoked in routine editing contexts where the user did not explicitly ask for style obfuscation. In practice, that can cause unintended rewriting of text, including removal of disclosure language, provenance markers, or compliance wording, which is risky because the skill is specifically designed to make AI-written text appear more human.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal