video-use-best-practices

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads canonical Hebrew fonts (Heebo, Rubik, Assistant) and python-bidi for text shaping. These are standard open-source resources used as intended for the skill's purpose.
  • [COMMAND_EXECUTION]: The skill uses ffmpeg, ffprobe, and fc-list to process video and verify font installation. All shell commands are scoped to media processing and environment verification within the project directory.
  • [REMOTE_CODE_EXECUTION]: The skill contains instructions for installing common Python packages (python-bidi) and binary tools (ffmpeg static builds) required for its core functionality. These are well-known, trusted tools for media workflows.
  • [DATA_EXPOSURE]: The skill reads environment variables (specifically ELEVENLABS_API_KEY) to facilitate ElevenLabs Scribe API calls. This is a standard practice for API-driven agents and is not used for exfiltration.
  • [PROMPT_INJECTION]: The skill includes instructions to provide users with pricing disclaimers and cost estimates before initiating paid API services, which is a defensive measure to protect the user from unexpected costs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 02:20 PM
Security Audit — agent-trust-hub — video-use-best-practices