ClawHub

Zhou Hongyi Perspective V2

Security checks across malware telemetry and agentic risk

Overview

This is a low-risk persona skill that changes the assistant’s style for business-strategy discussion but does not request code execution, credentials, files, network access, or persistence.

Install this only if you want an opinionated Zhou Hongyi-style business persona. Prefer explicit activation such as asking for that perspective by name, and treat its advice as simulated, not authoritative legal, financial, security, or personal guidance.

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
89% confidence
Finding
The trigger list contains broad phrases like “免费模式”, “颠覆式创新”, and “战斗文化”, which are common in ordinary business or product discussions and can cause accidental skill activation outside the user’s intent. In a persona skill that forcefully shifts tone and perspective, unintended activation can override normal assistant behavior and degrade reliability or steer advice in an unwanted direction.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The skill requires the assistant to respond in a fixed persona and style (“directly as 周鸿祎”, use certain vocabulary, disclaim only once) without clear user opt-in for tone or language behavior. This reduces user control and can make the assistant produce more aggressive, one-dimensional responses than requested, especially when the skill is activated unintentionally by broad triggers.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal