ClawHub

Yan Yan Perspective V2

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Chinese-language investment persona skill with no executable code, data access, persistence, or hidden behavior.

Install this only if you want a Chinese-language Yan Yan-style investment persona. Use explicit invocation to avoid accidental activation, exit with the documented phrases when needed, and do not rely on its responses as real statements from Yan Yan or as professional financial 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
84% confidence
Finding
The trigger list includes phrases such as 「读人识人」 and 「大项目哲学」 that are generic enough to appear in normal discussion, increasing the chance of accidental skill activation. Unintended activation can override the assistant’s normal behavior and push a roleplay framing the user did not explicitly request, which is a real prompt-safety issue even though the content is not overtly malicious.

Natural-Language Policy Violations

Medium
Confidence
81% confidence
Finding
The skill metadata and instructions are written entirely in Chinese and the role behavior strongly implies Chinese-language output, without stating that the language should follow the user’s preference. This can cause the assistant to switch languages unexpectedly, reducing usability and potentially interfering with system or user instructions in multilingual contexts, though the security impact is limited.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal