ClawHub
Back to skill

Security audit

Chen Humanizer

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed text-editing skill that rewrites AI-sounding prose, with no executable code, persistence, network payloads, or hidden data access found.

Install only if you want an editing aid that rewrites text for a more natural style. Do not use it to hide AI assistance where attribution, academic rules, employment rules, platform policies, or legal duties require disclosure, and review edits before applying them to important documents.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill is explicitly designed to make AI-generated text appear more human-written, but the description contains no warning against deceptive, fraudulent, or policy-sensitive use. In context, that omission is significant because the tool can be used to evade AI-detection heuristics, obscure disclosure obligations, or misrepresent authorship in academic, professional, or platform-governed settings.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The README says to 'invoke directly when editing documents,' which is broad guidance that can cause the skill to activate in many contexts without clear user intent or scope limits. Because this skill rewrites text specifically to remove signs of AI authorship, unintended activation could alter content in sensitive workflows and facilitate deceptive presentation of machine-generated text.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill explicitly instructs users to replace curly quotation marks with straight quotes, framing one punctuation convention as inherently more 'human.' This encodes an unstated language/locale preference and may degrade valid regional writing styles without offering a choice or limiting the skill to a specific English-house style.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.