The Agent Skills Directory

[COMMAND_EXECUTION]: Several scripts, including create_section_video.py and slides_to_video.py, utilize the subprocess.run and shutil.which functions to interface with system-level tools such as ffmpeg and ffprobe. These commands are used to merge audio files, generate video segments from images, and concatenate clips into a final presentation. The logic includes sanitization via slugification for filenames derived from slide headers, reducing the risk of command injection.\n- [EXTERNAL_DOWNLOADS]: The skill makes network requests to the OpenAI API for audio narration generation and uses the Google GenAI SDK to interact with the Gemini API for slide image generation. These external interactions are directed toward official, well-known technology service providers and are strictly limited to the stated purpose of generating presentation assets.\n- [DATA_EXFILTRATION]: User-provided content from markdown files, including speaker notes and slide text, is transmitted to OpenAI and Google Gemini. This data transfer is essential for the skill's core functionality (narrated video generation) and is secured using standard environment variables for authentication with the respective services.

markdown-video