ClawHub
Back to skill

Security audit

Anti AI Slop Writing Gate

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed writing-editing helper with no executable code, persistence, credential access, or external data flow, though users should know its pre-send mode silently polishes substantive replies.

Install this if you want an automatic prose-polishing gate. Use detect or edit mode instead of pre-send when you need visible change notes, auditability, or approval before wording changes are applied to important communications.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly enables a 'pre-send' mode that silently inspects and alters substantive replies before delivery, without notifying the user. Even though the goal is stylistic improvement, hidden modification of outputs can override user intent, introduce unreviewed changes, and reduce transparency and auditability in consequential communications.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.