ClawHub
Back to skill

Security audit

Audio Cog

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only audio-generation skill whose external API use and cloned-voice capability are disclosed, though users should apply consent and privacy care for voice cloning.

Install only if you trust CellCog and are comfortable sending prompts, scripts, and audio-generation requests to its service and providers. Use a controlled API key, monitor usage, avoid including secrets in prompts, and create or use cloned voices only with clear authorization from the voice owner and disclosure when sharing synthetic audio.

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
95% confidence
Finding
The skill explicitly promotes cloned-voice generation and avatar-based speech without any warning or guardrails around consent, impersonation, identity misuse, or handling of biometric voice data. In an audio-generation skill, this omission materially increases the risk that users will create deceptive or non-consensual voice content, including fraud, harassment, or privacy violations.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.