The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill performs intended media generation tasks using established service providers and standard library practices.\n- [COMMAND_EXECUTION]: The skill uses FFmpeg for video processing. Analysis of generate_video.py, add_voiceover.py, and images_to_video.py confirms that shell commands are executed using argument lists, which is a secure method that avoids shell injection vulnerabilities. The use of the concat demuxer with -safe 0 is restricted to file lists generated by the script itself.\n- [EXTERNAL_DOWNLOADS]: The skill downloads media assets from established AI providers (OpenAI, LumaAI, Replicate) as part of its core functionality. These operations target official API domains and are necessary for the skill's purpose.\n- [CREDENTIALS_UNSAFE]: The skill implements safe credential handling by instructing users to use environment variables or .env files for API keys, rather than hardcoding sensitive information.

ai-video-gen