ClawHub
Back to skill

Security audit

优质投资书单推荐

Security checks across malware telemetry and agentic risk

Overview

This appears to be a low-risk book recommendation skill with some usability and routing quirks, not a skill that seeks sensitive access or hidden control.

Safe to install for book recommendations, especially if you want Chinese-language output. Be aware it may activate on broad investing or business mentions and may answer in Chinese even when another language would be more appropriate.

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 skill’s trigger scope is overly broad: it activates not only for explicit book-recommendation requests, but also for general mentions of investing, business, economics, long-term thinking, or even a specific site. This can cause unintended routing of ordinary conversations into a constrained skill that forces a single-source workflow, increasing the chance of irrelevant responses, user-intent mismatch, and unnecessary external access.

Natural-Language Policy Violations

Medium
Confidence
86% confidence
Finding
The skill mandates Chinese output without checking the user’s language preference. This can produce unusable or confusing responses for users asking in other languages, and may lead to accidental disclosure or mishandling when multilingual users expect answers in the input language.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.