ClawHub
Back to skill

Security audit

📝 英语语法突破教练

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only English grammar coaching skill whose behavior is disclosed and aligned with teaching, with only minor user-control caveats.

Before installing, expect this skill to favor guided questions over direct corrections and to offer ongoing learning tracking only when you consent. Use it when you want grammar coaching, not quick proofreading.

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
92% confidence
Finding
The skill repeatedly says grammar-related scenarios should '务必调用' this skill, creating overly broad activation criteria that can trigger it in many general English-learning contexts. Over-broad invocation can override user intent or route conversations into unnecessary data collection and a rigid teaching flow, reducing user control and increasing the chance of inappropriate activation.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The skill enforces a single interaction style by stating it should always question and never directly give answers, without requiring user opt-in at activation time. This can deny users appropriate assistance, frustrate accessibility needs, and create unsafe or low-quality behavior when the user needs a direct explanation rather than Socratic coaching.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.