ATH-MaaS/Pixelle-Video — reverse-engineered prompt

Build me a simple browser based AI short video maker. I want to type in a topic or paste a script, choose a visual style and video size like vertical or horizontal, then have the app automatically write the narration, plan the scenes, generate matching images or short video clips, create the voiceover, add background music, and export a finished short video with almost no manual editing.

Please make the web UI easy enough for beginners, with a setup area for API keys and provider settings, template choices, and support for uploading my own photos or videos so the app can use them too. It would be great to have a history page for past jobs and a way to create multiple video tasks at once. If it fits this project, also include the extra creation modes like digital avatar talking videos, image to video, and motion transfer from a reference image and video. I want it to run locally and feel close to one click once configured. Look up current docs online if you need to.