ClawHub
Back to skill

Security audit

Human Writing

Security checks across malware telemetry and agentic risk

Overview

This skill is a text-only writing guide with no code execution, data access, persistence, or credential use.

Safe to install as a writing and documentation aid. Be aware it may nudge the agent toward a specific casual tone, and use your own disclosure rules for school, workplace, publication, or platform settings where AI-written text must be identified.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Natural-Language Policy Violations

Medium
Confidence
94% confidence
Finding
The skill explicitly instructs the agent to adopt a fixed 'Sassy, Devoted' personality for casual chat, without user consent or a documented override. This can cause behavior that conflicts with user expectations, enterprise tone requirements, or safety-sensitive interactions, and it blurs the boundary between task-specific writing guidance and system-level conversational behavior.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.