Why we built this

If you’ve ever watched an AI-generated video and wondered “what prompt made this?”, you’ve probably run into the same problem we did: most tools that claim to answer that question just guess — and present the guess as if it were fact.

VideoPromptGen exists to fix that one specific thing. We check whether a video’s creator actually shared their prompt first. If they did, we show you their exact words. If they didn’t, we tell you clearly that what you’re looking at is a reconstruction — an AI’s best-effort description of what’s visible in the video, not a recovered original.

That distinction is the whole point of this tool. No video-to-prompt tool, ours included, can pull an exact original prompt out of pixels alone. We’d rather say that plainly than pretend otherwise.

What the tool actually does

• Paste a YouTube link, and we check the video’s description and comments for a prompt the creator disclosed themselves.
• No disclosed prompt? An AI vision model studies the video and writes a best-effort reconstruction — clearly labeled as a guess, describing only what’s genuinely visible (framing, subject, color, mood, setting), never inventing technical details like camera or lens that can’t actually be seen in a still frame.
• Upload your own video instead, if you’d rather not use a YouTube link. Frames are extracted directly in your browser — the video file itself is never sent to our servers.
• Reconstructed results come phrased three ways, matching the conventions of Sora, Kling, and Runway, so you can take the description straight into whichever tool you’re using.

Who’s behind VideoPromptGen

VideoPromptGen is built and operated by H-Chuks, a registered business based in Oulu, Finland.

Questions or feedback?

We’d genuinely like to hear from you — whether it’s a bug, a suggestion, or just a question about how something works. Visit our Contact page to get in touch.