Security audit
voice-chat-mode
Security checks across malware telemetry and agentic risk
Overview
This is a text-only Chinese voice-mode style skill that does not install code or request secrets, so it looks safe to install with ordinary caution.
Before installing, make sure you want a persistent Chinese spoken-output style mode and know how to disable it. Host-level safety rules should still take priority if a prompt requires a refusal or warning.
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
VirusTotal
64/64 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.