ClawHub
Back to skill

Security audit

anti-ai-writing-patterns

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed writing-style skill with no executable code, data access, persistence, or privileged behavior, though its style rules are broad.

Install this only if you want the agent to consistently avoid common AI-writing patterns. Expect it to influence general replies, not just editing tasks, so disable or override it when you need a specific tone, legal/medical/security wording, or exact user-provided formatting preserved.

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
95% confidence
Finding
The skill explicitly states that its writing rules apply to all output regardless of model, context, or language, which overrides normal task-specific behavior and user preferences. That creates a cross-context policy injection risk: it can silently reshape security analyses, legal text, medical text, or user-requested tone/format in ways the user did not authorize.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.