ClawHub

De-AI-fy Text Skill1.0.0

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed local text-rewriting skill with ethical misuse concerns, but no evidence of hidden access, exfiltration, persistence, or unsafe automatic behavior.

Install only if you are comfortable with the publisher and the authenticity implications. Use it on copies or dedicated folders, review outputs manually, and do not use it where AI assistance must be disclosed, such as academic, professional, moderation, or authorship-sensitive settings.

SkillSpector

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

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The README explicitly markets a tool for removing AI-generated text characteristics and making output appear more human-written, but it does not warn about deception, impersonation, academic dishonesty, spam, or other authenticity-related misuse. In this context, the omission is security-relevant because the skill is purpose-built to reduce detection signals and increase plausibility of synthetic text, which can facilitate social engineering and policy evasion.

Ssd 2

Medium
Confidence
98% confidence
Finding
This code explicitly markets and implements functionality to remove AI-like features and make generated text appear more human. In context, that is an evasion capability that can be used to bypass AI-content detection, authenticity checks, academic or platform policy enforcement, and other trust/safety controls.

Ssd 4

Medium
Confidence
96% confidence
Finding
The staged rewriting pipeline applies multiple probabilistic transformations—transition replacement, formal-word replacement, sentence-pattern rewriting, and colloquialization—to gradually alter text while preserving meaning. In this skill's context, that layered design increases its effectiveness as a stealth paraphrasing/evasion tool, making deceptive output harder for downstream detectors and reviewers to identify.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal