audio-to-subtitles

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads and executes scripts using npx -y bun, which is a standard execution pattern for Node.js-based tools.
  • [COMMAND_EXECUTION]: The skill uses npx and bun to execute its own local scripts as demonstrated in the documentation. This is part of its core functionality for processing media files.
  • [DATA_EXFILTRATION]: The skill uploads local audio/video files to Cloudflare R2 to provide a public URL for the transcription API. This is a documented and necessary step for the Volcengine AI MediaKit ASR service, which requires public URLs. The credentials for R2 are managed via environment variables.
  • [CREDENTIALS_UNSAFE]: The skill requires API keys for MiniMax, Volcengine, and Cloudflare R2. The implementation correctly retrieves these from environment variables or a local .env file and does not hardcode any secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 12:53 AM
Security Audit — agent-trust-hub — audio-to-subtitles