ClawHub
Back to skill

Security audit

来怼

Security checks across malware telemetry and agentic risk

Overview

This is a text-only Chinese critique skill that pressure-tests ideas and does not request system access, credentials, background execution, or data access.

Install this if you want blunt Chinese-language pressure testing of your own ideas. Be careful using it for health, legal, financial, or major personal decisions, because it is a critique framework rather than qualified professional advice.

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 trigger scope is intentionally very broad ('不限项目、不限领域') and matches many ordinary requests for critique about ideas, decisions, or plans. That can cause over-activation and unintended interception of normal conversations, steering users into an aggressive red-team mode without clear consent or domain boundaries, especially in sensitive contexts like health, legal, finance, or personal decisions.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
Mandating Chinese output regardless of user language can create misunderstanding, reduce transparency, and impair safe use if the user cannot fully understand critiques or caveats. In a skill designed to pressure-test ideas across any domain, language mismatch increases the chance of misinterpretation, especially when discussing consequential topics.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.