ClawHub
Back to skill

Security audit

课堂互动教练

Security checks across malware telemetry and agentic risk

Overview

This is a teacher-facing classroom interaction planning skill with disclosed, purpose-aligned use of lesson context and aggregate classroom observations.

Installers should treat this as a classroom planning aid, not an automated student-monitoring tool. Teachers should avoid recording student names or sensitive individual performance details unless their school policy permits it, and should confirm intent before using the skill on very short prompts like “怎么分组” or “活跃度低.”

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
88% confidence
Finding
The activation guidance uses broad natural-language examples such as '这节课怎么互动' and related variants without defining exclusions, which can cause the skill to be invoked in loosely related teaching conversations. In an agent environment, over-broad routing can expose lesson-planner and student-analyzer data flows unnecessarily and produce unintended behavior when the user did not actually request this capability.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger table contains everyday phrases like '课堂冷场', '活跃度低', and '怎么分组' that are semantically broad and lack boundary conditions, making accidental invocation likely. Because this skill coordinates with other skills and writes aggregated interaction logs, mistaken activation can lead to unnecessary processing of classroom or student-related context and reduce user control over which tool is used.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.