video-generation-minimax

What problem does it solve?

This Skill solves the problem of producing short videos from ideas and reference materials without manually editing footage by automating calls to MiniMax’s video generation APIs.

Core Features & Use Cases

• Text-to-Video Generation: Turn a natural-language prompt into a generated video for quick ideation and storyboard drafts.
• Image-to-Video & Start/End Frames: Create motion by using an initial image (or both first and last images) plus a prompt for scene continuity.
• Subject Reference (Face Consistency): Generate videos based on a character/subject image and prompt to keep the subject’s features consistent.
• Asynchronous Polling + Download: Automatically submit a generation task, poll task status, retrieve the resulting file, and save it to your workspace.
• Use Case: Provide a prompt like “a woman in a coffee shop looking out the window” to generate a cinematic short clip for marketing or creative prototyping.

Quick Start

Use the skill to create a text-to-video by running: python3 {baseDir}/scripts/video_gen.py --mode text --prompt "镜头拍摄一个女性坐在咖啡馆里，女人抬头看着窗外".