ClawHub

zhujue-characters

Security checks across malware telemetry and agentic risk

Overview

This is a coherent local literary lookup skill, but it can surface mature or discriminatory passages from the bundled novel and should be used with content-aware handling.

Install only if you want a local reference tool for this specific novel. Be aware it bundles the full text and may retrieve sensitive passages; use content warnings, summarize instead of quoting harmful material by default, and avoid exporting large chapter sets unless needed.

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 (2)

Natural-Language Policy Violations

Medium
Confidence
97% confidence
Finding
The content repeatedly uses derogatory and stigmatizing language about gender expression and perceived sexual orientation without any safety framing, transformation, or user-controlled filtering. Even though this appears in literary source text rather than as an instruction, importing and presenting it unmediated in a skill can propagate harmful slurs and hostile characterization to end users.

Natural-Language Policy Violations

Medium
Confidence
93% confidence
Finding
This file contains repeated, explicit suicide and self-harm content, including method and aftermath details, without any visible warning, age-gating, or safety handling. In a retrieval skill that imports raw novel chapters for downstream use, this can surface graphic self-harm content unexpectedly to end users or other agent components, increasing harm risk for vulnerable users and creating unsafe completions if the text is quoted or summarized without safeguards.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal