omniroute-stt

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation references a configuration file located on a public GitHub repository at https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/omniroute/SKILL.md. This is a reference to a whitelisted domain for documentation purposes.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to transcribe audio data into text, which creates an inherent attack surface where spoken instructions in an audio file could attempt to influence the agent's behavior.
  • Ingestion points: Audio files (e.g., audio.mp3) uploaded via the transcriptions and translations endpoints.
  • Boundary markers: The instructions do not define specific delimiters (like XML tags or triple quotes) to isolate the transcribed text from the rest of the agent prompt.
  • Capability inventory: The skill documentation includes examples using curl for network requests and file system referencing (-F "file=@audio.mp3").
  • Sanitization: The skill does not provide instructions for sanitizing, filtering, or validating the output of the transcription before it is interpreted by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 07:22 PM