ZeroLu/Ultimate-AI-Media-Generator-Skill — reverse-engineered prompt

Build me an open source AI media generator skill that lets coding agents like Cursor, Claude Code, Codex, and OpenClaw create images and videos for me without me needing to write perfect prompts. It should connect to CyberBara with an API key, save the key locally after setup, show my credit balance and recent usage, and let me ask for a price quote before starting a generation.

I want simple chat style examples for making social post images, product hero images, short promo videos, slide visuals, SEO article images, and comic drama scenes. Support text to image, image to image, text to video, image to video, and video to video where the model allows it. Include popular options like Nano Banana, Sora, Seedance, Kling, and Veo, with a sensible fallback if one model is not available.

Please include clear install steps, command line helpers, reusable workflow templates, and a curated prompt library. Look up current docs online if you need to.