ClawHub
Back to skill

Security audit

升学规划助手

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent college-admissions planning helper with local reference data and small offline analysis scripts; its optional API-key mode is disclosed.

Safe to install for offline admissions guidance. Only enable commercial/API mode if you trust the configured endpoint and provider, avoid sharing unnecessary personal identity or contact details, and verify final admissions decisions against official provincial education-exam sources.

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 (1)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger is described broadly enough that the skill may activate for general gaokao-related conversation, causing unintended routing and over-application of the skill. This is not a direct exploit primitive, but it can create security and safety issues by bypassing more appropriate skills or injecting domain-specific guidance into casual discussion.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal