elevenlabs-stt

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the infsh CLI to perform audio transcription and alignment. This access is explicitly restricted in the allowed-tools configuration to ensure the agent only executes authorized vendor commands.
  • [EXTERNAL_DOWNLOADS]: Provides a link to installation instructions and setup scripts located on the author's official GitHub repository (inference-sh/skills).
  • [DATA_EXFILTRATION]: Audio data is processed by passing URLs to the infsh tool. No sensitive local files or credentials are accessed or transmitted externally.
  • [PROMPT_INJECTION]: The instructions focus entirely on tool usage and parameter configuration without any patterns designed to override agent safety or behavioral guidelines.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 09:05 PM