openai-whisper-api

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a bash script (scripts/transcribe.sh) to execute curl commands. The script follows best practices with set -euo pipefail and handles user-provided arguments safely without dangerous shell interpolation.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with OpenAI's official API endpoint (api.openai.com) to perform transcriptions. This is a well-known service, and no executable code or scripts are downloaded from remote sources.
  • [DATA_EXFILTRATION]: The skill sends audio data to OpenAI's servers as intended by its primary function. It correctly manages the OPENAI_API_KEY via environment variables rather than hardcoding credentials, which is a standard and secure practice.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 12:14 AM