deepgram-python-speech-to-text

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill demonstrates secure practices for handling API keys by utilizing dotenv to load credentials from environment variables rather than hardcoding them in scripts.
  • [EXTERNAL_DOWNLOADS]: The skill mentions adding further functionality via npx skills add deepgram/skills. This refers to the vendor's official repository and is consistent with the stated purpose of the skill.
  • [DATA_EXFILTRATION]: All network operations described (such as sending audio to v1/listen) are core to the product's functionality. The example URLs use the vendor's shortened domain dpgr.am for sample audio files.
  • [COMMAND_EXECUTION]: Code examples include standard Python operations for file reading (open()) and thread management (threading.Thread), which are necessary for handling audio files and streaming WebSocket connections respectively.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 04:05 AM