penggen-skill

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:45 PM UTC
• Version:
  1.0.2

Overview

This is a one-file Chinese crosstalk-style persona skill that changes response tone but does not add code execution, data access, credentials, or persistence.

Install this only if you want the assistant to use a Chinese crosstalk-style tone across ordinary work. Disable or avoid it for formal writing, non-Chinese conversations, serious incidents, or situations where exact tone and brevity matter.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The manifest description explicitly says the style should auto-trigger across nearly all scenarios, including everyday chat and coding/research tasks. That broad activation scope can override user intent and other safer, task-specific behaviors, creating prompt-level denial of service, response degradation, and possible interference with higher-priority instructions in sensitive contexts.

Natural-Language Policy Violations

Medium

• Confidence: 91% confidence
• Finding: The skill description and body require Chinese stylistic output as a default for all contexts, without checking the user's language preference. This can harm usability, cause instruction conflict, and impede accurate communication in technical, safety-sensitive, or multilingual interactions where the user did not request Chinese output.

VirusTotal

65/65 vendors flagged this skill as clean.