ClawHub
Back to skill

Security audit

mano-asr

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only skill for local speech-to-text integration; its main risks are optional privacy-impacting settings users should understand before enabling.

Use the minimal local-only configuration if privacy is important. Only enable OpenAI/cloud fallback if you are comfortable sending audio to that provider, and only enable transcript echo in trusted/private channels where storing or displaying the transcript is acceptable. Review the third-party Homebrew tap before installing in managed environments.

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 (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The document presents a 'local-first + cloud fallback' transcription configuration using OpenAI, but the surrounding guide strongly emphasizes offline privacy and zero-cost local processing. Without an explicit warning that fallback will transmit user audio off-device to a third party, operators may enable it without realizing the privacy boundary changes, causing sensitive voice content to leave the host unexpectedly.

Ssd 3

Medium
Confidence
84% confidence
Finding
The example enables transcript echoing back to the user, which can expose sensitive spoken content in plain text in chat history, notifications, logs, screenshots, and shared devices. This is especially risky because voice messages often contain passwords, personal data, financial details, or information spoken in environments where the user expects less durable storage than a text transcript.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.