ClawHub
Back to skill

Security audit

Discord语音免费

Security checks across malware telemetry and agentic risk

Overview

This skill is a basic Discord voice assistant whose sensitive behaviors are mostly disclosed and aligned with its purpose, but users should understand the voice-channel privacy and OpenAI TTS data-flow implications.

Install only in Discord servers and voice channels where participants understand that speech may be captured, transcribed locally, processed by the agent, and spoken back through OpenAI TTS. Use a dedicated Discord bot token and OpenAI API key, avoid sensitive conversations, and remove the bot from channels when not in use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The skill repeatedly claims the free version uses local Whisper without external API keys, yet the dependency section instructs users to export a generic API_KEY and references external API usage. This inconsistency can mislead operators into supplying unnecessary secrets, increasing the chance of credential exposure, misuse, or accidental routing of data to external services they did not expect.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly states that allowedUsers is not supported, meaning any user in the voice channel may trigger recording, transcription, agent processing, and TTS playback. Without a prominent privacy and authorization warning, users and administrators may deploy it in shared channels without understanding that uninvolved participants can be recorded or influence outputs.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The description highlights local Whisper and 'offline' behavior, but also includes OpenAI TTS without clearly warning that text content will be sent to an external provider for synthesis. This can create a false impression of end-to-end local processing and lead to unintentional disclosure of transcribed or generated content to a third party.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.