ClawHub

MLX TTS

Security checks across malware telemetry and agentic risk

Overview

This is a coherent local text-to-speech skill, with the main cautions that setup installs external tools and broad triggers may invoke it more easily than expected.

Install this only if you are comfortable allowing setup to install or update ffmpeg, uv, and mlx-audio from external package sources. Prefer explicit invocations like /mlx-tts when using it, and remember that text sent to the skill is rendered into an audio file and sent back through the message tool.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger set includes very generic phrases like "Say <text>", "TTS ...", and "Reply with voice message ...", which can overlap with normal conversation and common assistant requests. This increases the chance the skill is invoked unintentionally or preempts other safer/more appropriate skills, causing unexpected local command execution and message sending behavior.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The installer automatically invokes Homebrew and uv to install software without any prompt, warning, or validation. This can cause unintended system changes and executes code fetched from external package sources, which is risky in an agent-skill context where users may not expect installation side effects during setup.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal