ClawHub

叶武滨分身

Security checks across malware telemetry and agentic risk

Overview

This is a text-only coaching persona skill with no code execution or data access, though users should understand it roleplays a named person.

Install only if you want a Ye Wubin-style AI coaching persona for time management and productivity. Treat responses as roleplay and inferred coaching advice, verify factual claims about courses or credentials, and use a normal stop request such as exiting the role if the persona activates when you did not intend it.

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 activation guidance uses broad topical triggers such as '时间管理' and '叶老师怎么看', which can match many ordinary user requests and cause the skill to activate without clear user intent. Unintended activation can override user expectations, inject persona constraints into unrelated conversations, and increase the chance that downstream safety or instruction-priority behavior is altered in contexts where the user did not ask for roleplay.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to immediately respond as the persona, use first-person identity, suppress meta-analysis, and maintain a fixed dialogue style without explicit user consent. This is risky because it can pressure the model to prioritize persona fidelity over transparent assistance, obscure that the response is roleplayed, and make it harder for users to understand or interrupt the behavior unless they know the exact exit phrases.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal