NYC

speech-to-text

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill recommends installing external packages including 'elevenlabs', 'python-dotenv', 'pydub', and various '@elevenlabs/*' npm modules to support its transcription features.
  • [DATA_EXFILTRATION] (LOW): The skill performs network operations to 'api.elevenlabs.io'. While necessary for the transcription service, this targets a domain outside the predefined trusted whitelist.
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. Transcribed text from external audio sources could potentially contain instructions that influence the agent's behavior.
  • Ingestion points: Audio and video files provided to the 'convert' method in 'SKILL.md' and streaming URLs used in 'references/realtime-server-side.md'.
  • Boundary markers: Absent. The transcribed text is returned to the agent without specific delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill possesses the capability to read local audio files and communicate with the ElevenLabs API over the network.
  • Sanitization: No sanitization or filtering of the transcribed text is performed before it is integrated into the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:33 PM