This is an automated video creation agent built on LLM/VLM. Users can describe their intentions in natural language, and the system automatically accomplishes material collection, editing, storyline construction, voiceover, and background music creation. It supports conversational refinement, including cutting, reorganizing, modifying subtitles, etc., and can be saved as a clip skill pack with one click for quick reproduction of videos in the same style