The Agent Skills Directory

[SAFE]: The skill demonstrates secure coding practices. The scripts/sop.py utility validates all asset paths using a robust check to prevent directory traversal and ensure files remain within the project scope. All media processing via ffprobe and ffmpeg uses safe argument passing with validated inputs.\n- [SAFE]: External dependencies are managed securely. The skill references the GSAP library via a trusted public CDN and uses npx to fetch a specific, pinned version of the hyperframes renderer from a well-known service provider.\n- [SAFE]: The potential for indirect prompt injection via documentation or screenshot ingestion is addressed through standard mitigations. The build.mjs script performs HTML escaping on all user-supplied titles and instructions during HTML generation to prevent injection in the rendered output.\n- [SAFE]: No hardcoded credentials or sensitive information were found. The documentation provides clear guidance on managing API keys for external voice services through local environment variables, adhering to secret management best practices.

sop-video