krillinai/KrillinAI — reverse-engineered prompt

Build me a simple AI video translation and dubbing tool for localizing videos from one language to another. I want to be able to paste a video link or upload a local file, then have it handle the full flow, download if needed, transcribe speech, split subtitles cleanly, translate them naturally, generate dubbed audio, burn in or export subtitles, and render the final video. It should work well for both landscape and vertical videos so content can be reused for places like YouTube, TikTok, Bilibili, and similar short video platforms.

Please make it friendly for normal users with a one click experience, but also include command line stages so each step can run on its own for automation or agent workflows. I also want optional cover image generation from the video thumbnail and prompt based text, terminology replacement, and support for different speech recognition, translation, and voice providers through standard API style configs. It should run on Windows, Mac, and Linux. If anything is unclear, check the current docs online and wire up a sensible default config.