ClawHub
Back to skill

Security audit

Li Kaifu Perspective V2

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only Chinese persona skill for Li Kaifu-style analysis, with no evidence of code execution, credential access, persistence, or data exfiltration.

Install this only if you want a Chinese-language analytical persona for AI, startup, career, and work-health questions. Do not present its output as Li Kaifu's actual views or direct quotes, and verify factual claims or investment advice before acting on them.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
95% confidence
Finding
The skill auto-activates on broad scenario phrases like investment evaluation, tech trends, and work-health balance, which can overlap with ordinary user requests that do not explicitly ask for this persona. That can cause unintended skill takeover, steering responses into a specific viewpoint and reducing user control over how the assistant interprets ambiguous prompts.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The skill is written to strongly enforce Chinese-language output and persona behavior across the entire file, without any indication that the user's language preference should take priority. This can override user expectations, degrade usability for non-Chinese requests, and make the assistant less reliable by privileging skill instructions over the active conversation context.

Static analysis

No suspicious patterns detected.