voice-assistant

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it transcribes user-provided audio files and makes the text available to the agent context.
  • Ingestion points: Audio file processing via the whisper command in SKILL.md.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are used during text processing.
  • Capability inventory: The skill uses curl for network requests and system audio tools (afplay/aplay).
  • Sanitization: No sanitization of the transcribed text is mentioned before it is processed by the agent.
  • [COMMAND_EXECUTION]: The skill executes local system commands including ffmpeg, whisper, afplay, and aplay for media conversion and playback, which are necessary for its primary functionality.
  • [SAFE]: The skill makes network requests to Mistral AI's official API (api.mistral.ai) for text-to-speech generation, which is a well-known service. It uses environment variables for secure API key management.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 07:16 AM
Security Audit — agent-trust-hub — voice-assistant