Humanize Chinese 2.0.0

• Outcome: Review
• Latest audit: May 29, 2026 at 9:24 PM UTC
• Version:
  1.0.0

Overview

This is a local Chinese text rewriting tool, but it is explicitly marketed and designed to reduce AI-detection signals.

Install only for legitimate editing and style-improvement work. Do not use it to bypass academic, workplace, publisher, or platform AI-disclosure rules, and review outputs before replacing originals.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The README explicitly markets the tool as making AI-generated Chinese text 'natural and undetectable,' which facilitates concealment of AI authorship and deceptive content transformation. Even though this is documentation rather than executable code, the skill’s stated purpose materially enables misuse by encouraging evasion of detection and omits any safety boundaries or acceptable-use guidance.

Natural-Language Policy Violations

High

• Confidence: 97% confidence
• Finding: The phrase 'undetectable' directly signals an evasion objective, which conflicts with policies against concealment of AI involvement and increases the likelihood that the skill will be used to bypass reviewers, platforms, educators, or trust-and-safety controls. In context, the README reinforces this risk with features such as detection scoring, sentence-level analysis, and humanization workflows that appear optimized to identify and remove machine-like signals.

VirusTotal

61/61 vendors flagged this skill as clean.