speech-to-text

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes documentation and examples for using curl to interact with the Each Labs API endpoint at https://eachsense-agent.core.eachlabs.run/v1/chat/completions. This is the intended mechanism for the transcription service.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the processing of audio and video content hosted at external URLs provided by the user to the transcription API.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it processes untrusted audio and video content from external URLs which could contain malicious verbal instructions.
  • Ingestion points: External URLs for audio and video files in the image_urls parameter in SKILL.md.
  • Boundary markers: Absent.
  • Capability inventory: Use of Bash(curl) and WebFetch for API interaction.
  • Sanitization: No explicit sanitization or filtering of transcribed content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 12:03 PM
Security Audit — agent-trust-hub — speech-to-text