Security audit

ElevenLabs Voice Pipeline

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it says, but voice cloning and Telegram delivery involve sensitive voice and message data without enough consent and privacy guardrails.

Review before installing. Use voice cloning only for your own voice or for speakers who have given explicit informed permission, avoid submitting secrets or regulated personal data, and store ElevenLabs and Telegram credentials in environment variables or a secrets manager.

SkillSpector

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 (2)

Missing User Warnings

High

Confidence: 95% confidence
Finding: The skill includes detailed voice-cloning workflows and sample-upload handling but does not present any user-facing warning about consent, biometric sensitivity, impersonation risk, or third-party processing of voice data. Because voiceprints are sensitive biometric data and cloning can enable fraud or impersonation, omission of explicit safeguards materially increases misuse risk in this context.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The Telegram integration transmits generated audio to Telegram and source text to ElevenLabs without clearly warning users that their content leaves the local environment and is processed by external services. This can expose sensitive text or audio content unintentionally, especially in agent contexts where users may assume local-only handling.

Static analysis

No suspicious patterns detected.