Emotion switch

Security checks across malware telemetry and agentic risk

Overview

This is a text-only style skill that changes the assistant’s conversational emotion tone and does not add code execution, data access, or external connections.

Install this only if you want a Chinese-oriented conversational style skill that can keep the assistant in a chosen emotional tone for a chat. Avoid or constrain it where strict language mirroring, neutral tone, or explicit disclosure of style modes is required.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The skill explicitly states that qiqing-liuyu's Chinese-language behavior is always in effect, which can override the user's preferred language and constrain responses without fresh user consent. This is a policy/UX security concern because it can cause instruction misalignment, reduce accessibility, and make the assistant ignore legitimate user requests for another language.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal