ClawHub
Back to skill

Security audit

Web Novel Writing Reference

Security checks across malware telemetry and agentic risk

Overview

This is mainly a web-novel writing reference skill, but it tells the agent to automatically install and load another skill when activated.

Review before installing. This does not look malicious, but users should only enable it if they are comfortable with it adding and loading Universal Task OS automatically; a safer version would ask first, identify the dependency source and version, and remain in read-only reference mode unless the user opts into UTOS-backed writing workflows.

SkillSpector

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

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill explicitly instructs the system to automatically install a dependency during activation without user confirmation. This creates an unauthorized capability expansion path: a seemingly read-only reference skill can cause code/content from another skill to be fetched and loaded, increasing supply-chain and prompt-surface risk without informed consent.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.