ClawHub

Humanize Text

Security checks across malware telemetry and agentic risk

Overview

This skill is a text-rewriting helper with a small local detection script and no evidence of hidden access, persistence, network activity, or destructive behavior.

Install it as a low-risk writing aid, but be aware it may trigger on broad Chinese copywriting requests and may rewrite tone more aggressively than intended; review outputs before using them in customer-facing or sensitive contexts.

Publisher note

检测和修复AI生成文本中常见的机械感、模板化表达、过度礼貌等问题，让文本更像人写的。适用于小红书文案、客服回复、产品介绍等场景，自带AI味检测脚本。

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
91% confidence
Finding
The trigger phrases are broad, generic user intents like '优化文案' and '自然一点' that could match many ordinary requests outside the intended skill scope. This can cause unintended skill activation, overriding a user's broader context or routing benign conversations into style-rewriting behavior unexpectedly.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal