speech-to-text

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions hosted on the inference-sh GitHub repository and suggests installing auxiliary tools using the platform's skill management command.
  • [COMMAND_EXECUTION]: The skill utilizes the 'belt' command-line interface to execute transcription tasks via Bash, which is the intended functional purpose of the integration.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes audio content from external URLs, which represents a potential attack surface if the transcribed text contains instructions that influence the agent's logic.
  • Ingestion points: Audio files retrieved via the audio_url parameter in CLI commands.
  • Boundary markers: None specified in the documentation.
  • Capability inventory: Execution of the 'belt' CLI tool for transcription and translation tasks.
  • Sanitization: No explicit content filtering or sanitization steps are described for the transcription output.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:18 PM
Security Audit — agent-trust-hub — speech-to-text