LiveKit Voice AI

Security checks across malware telemetry and agentic risk

Overview

This skill is a normal LiveKit voice-agent guide, with a caution that its phone-call example can contact real numbers if copied into a working project.

Before using this skill, protect API keys, review LiveKit and provider billing/privacy terms, and add safeguards before enabling outbound calls: use test numbers first, require explicit authorization and consent, restrict allowed destinations, rate-limit call creation, and keep audit logs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The skill documents code for placing outbound SIP calls to a real phone number but provides no warning about the real-world effect, consent requirements, cost, or abuse potential. In a voice-agent skill, this increases the chance that a user will copy and run telephony functionality that can contact external parties, incur charges, or violate policy/law if used without safeguards.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal