Security audit

Vn Write Guide

Security checks across malware telemetry and agentic risk

Overview

This is a non-executable writing guide, but it repeatedly gives explicit advice for bypassing NSFW/content filters.

Review carefully before installing. The skill does not appear to run code or access your files, but it normalizes bypassing content filters and recommends less-filtered providers for NSFW output; use it only within the rules of the platforms and services you rely on.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Context-Inappropriate Capability

High

Confidence: 99% confidence
Finding: This section explicitly claims that using specialized skills causes NSFW protections to be bypassed and presents that as an advantage over jailbreaks. That is not ordinary writing guidance; it is operational advice for defeating safety controls, increasing the likelihood the skill is used to generate disallowed sexual content or other moderated material.

Context-Inappropriate Capability

High

Confidence: 98% confidence
Finding: The model-selection table ranks providers by how well they allow NSFW content, whether moderation is lighter, and whether users should prefer API or 'Expert/Pro' variants to reduce filtering. This materially facilitates safety-evasion by directing users toward less-moderated services for restricted content generation.

Context-Inappropriate Capability

High

Confidence: 97% confidence
Finding: The troubleshooting guidance tells users to switch providers, choose weaker-moderated interfaces, or change wording when filters trigger. Those are direct workaround instructions for content moderation systems and exceed any legitimate need of a writing or translation guide.

Ssd 2

Medium

Confidence: 95% confidence
Finding: Advising users to substitute synonyms when keywords trigger filters is a textbook paraphrasing workaround for safety systems. This makes the skill more effective at slipping restricted requests past moderation and normalizes deliberate evasion behavior.

Ssd 1

High

Confidence: 99% confidence
Finding: The document explicitly frames specialized skills as superior to jailbreaks because they allegedly bypass NSFW protections with less 'attention noise.' This is dangerous because it repackages jailbreak intent into a procedural method for evading safety controls while preserving plausible legitimacy as a writing aid.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.