speech-to-text

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFE
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill requires an ELEVENLABS_API_KEY for authentication. It follows security best practices by encouraging the use of environment variables rather than hardcoding keys. In references/realtime-client-side.md, it explicitly warns against exposing API keys in client-side code and provides a secure architecture for generating single-use tokens on the backend.
  • [EXTERNAL_DOWNLOADS]: The skill references official ElevenLabs packages and APIs (e.g., api.elevenlabs.io, @elevenlabs/elevenlabs-js). These are verified vendor resources. It also includes examples of streaming audio from external URLs (e.g., NPR streams), which is a legitimate feature for a transcription service.
  • [DATA_EXFILTRATION]: Audio and video data are transmitted to ElevenLabs' infrastructure for processing. This behavior is documented, expected for the skill's primary purpose, and occurs over secure HTTPS/WSS connections.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 05:51 AM