skills/mbailey/voice-mcp/voicemode-dj/Gen Agent Trust Hub

voicemode-dj

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a custom command-line interface, voicemode dj, to execute media control tasks such as starting/stopping playback, adjusting volume, and searching for tracks. These commands interface with the mpv media player.
  • [EXTERNAL_DOWNLOADS]: The skill provides functionality to play media from external URLs and specifically integrates with remote music libraries (e.g., Music For Programming), which involves streaming data from external servers.
  • [PROMPT_INJECTION]: The voicemode dj play <url> command acts as a data ingestion point for untrusted external content. While the intended use is for audio streaming, this creates a surface for indirect prompt injection if the agent attempts to process or summarize metadata returned from external media streams.
  • Ingestion points: URL input in voicemode dj play <file|url> in SKILL.md.
  • Boundary markers: None present.
  • Capability inventory: Subprocess execution of voicemode dj CLI tools.
  • Sanitization: Not explicitly defined in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 03:26 AM
Security Audit — agent-trust-hub — voicemode-dj