ClawHub
Back to skill

Security audit

novel-writer

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed, instruction-only Chinese novel-writing skill that creates local project files for story planning and drafting, with no evidence of hidden execution, credential access, network use, or destructive behavior.

Install this only if you want a local-file-based Chinese novel-writing assistant. Be aware it includes explicit giantess and sexualized size-hierarchy worldbuilding templates, and it may create or update story project files, so use clear project names and avoid providing private reference text unless you are comfortable storing derived style notes locally.

SkillSpector

By NVIDIA
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 (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation criteria are broad enough to match generic fiction-related requests, which can cause the skill to trigger when the user did not intend to invoke a filesystem-writing novel project workflow. In this skill, unintended activation matters because the documented behavior includes creating directories, configs, outlines, and other project files, increasing the chance of surprising side effects.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documentation describes creating and updating project directories and files, but does not prominently warn that it will perform filesystem modifications. This can mislead users about the skill's side effects and, when combined with broad triggering, may result in unintended local file creation or overwriting within the workspace.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.