ClawHub
Back to skill

Security audit

书到技能 · 把技术书/PDF 蒸馏成 AI 技能

Security checks across malware telemetry and agentic risk

Overview

This skill coherently turns user-provided PDFs or documents into reusable skill packs, with the main caution being privacy, copyright, and authorization for the source documents.

Install only if you intend to process documents through your agent environment. Use documents you are authorized to transform, redact confidential or regulated content when needed, and review generated skills before importing or sharing them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The README encourages users to process books, PDFs, whitepapers, and internal manuals, but it does not warn that these inputs may contain sensitive company data, personal information, trade secrets, or copyrighted material that will be exposed to the AI processing workflow. In this skill context, that omission is more dangerous because the advertised use cases explicitly include internal operational manuals and proprietary knowledge, increasing the likelihood that users will upload confidential or restricted content without considering privacy, compliance, or IP risks.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The description contains very broad trigger phrases such as turning any book/PDF into a skill, which can cause the skill to activate in contexts the user did not intend. In an agent ecosystem, unintended invocation can lead to unnecessary file access, processing of sensitive documents, or confusing behavior where this skill overrides a more appropriate one.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The instruction to import via a specific Chinese-language UI path ('上传技能') assumes a locale without user opt-in, which can misdirect users or cause them to follow the wrong workflow. While not severe, forced locale behavior can reduce user control and increase the chance of mistakes during packaging or import steps.

VirusTotal

51/51 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.