social-persona-switcher

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:23 PM UTC
• Version:
  1.0.0

Overview

This is a text-only humor skill that rewrites a user’s sentence into exaggerated social personas and does not request code execution, credentials, network access, files, or persistence.

Install this only if you want an entertainment-style rewrite helper that may respond to ordinary sentences with exaggerated persona versions. Review outputs before posting or sending them in sensitive conversations.

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: 95% confidence
• Finding: The trigger condition is effectively 'any user sentence' plus several common phrasing examples, making the skill eligible to activate during ordinary conversation without strong user intent. In an agent environment, this can cause unintended persona-transformation behavior, override normal assistant responses, and create confusing or unsafe outputs when the user did not explicitly request the skill.

Natural-Language Policy Violations

Medium

• Confidence: 85% confidence
• Finding: The skill mandates automatic Chinese/English output selection based solely on input language rather than clear user choice. While not highly severe, forced output behavior can override user expectations, reduce transparency, and make the assistant behave in ways the user did not explicitly request, especially in multilingual contexts.

VirusTotal

66/66 vendors flagged this skill as clean.