ElevenLabs CLI

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed helper for using an unofficial ElevenLabs CLI, with expected privacy and supply-chain cautions but no artifact-backed hidden or malicious behavior.

Install only if you trust the community-maintained CLI and the package source you choose. Use a scoped or disposable ElevenLabs API key where possible, avoid sending sensitive audio/text unless you accept ElevenLabs processing it, be careful with voice cloning and public sharing, and use read-only or restricted modes if enabling MCP tools for an autonomous agent.

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)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill documents `voice share <voice-id>` with no warning about privacy, consent, or exposure implications. Because this skill supports voice cloning, sharing a cloned or sensitive voice publicly can expose biometric-like voice data or unauthorized likenesses, making the omission materially risky in context.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal