Skip to main content
Velo is the video layer over work you are already doing. You don’t start by making a video — you start with the work: a screen you are walking through, a recording that already exists, a document, a live URL, or just the thing you need to explain. Velo reads that context, writes the script, renders the video, and hands back one link to share and track. That is why there are five entry points rather than one. They are not five products; they are five kinds of work Velo can sit on top of.
Scene 1
The home screen groups these into three tabs. Start from scratch holds two cards — Start with a screen recording and Start with a prompt. Upload Video/PDF takes an existing file. Paste a link hands a URL to the browser agent.

​
Direct screen capture

The Capture screen option is designed for users who need to record a specific tab, window, or their entire desktop on the fly. This method uses the Velo browser extension to facilitate the recording process. It’s the preferred choice when you don’t have a pre-existing video file and want to begin automated processing immediately after finishing a live demonstration.

​
Uploading Recordings and Documents

For users who already have produced content, Upload a recording or PDF takes an existing file straight into processing. This method supports standard video formats including MP4 and MOV. Beyond video, this flow also supports uploading a PDF — specifically intended for sharing explanations of static documents such as pitch decks, PRDs, or project explainers. When a document is uploaded, the platform asks a few brief questions to help structure the resulting video message around the file’s content. Paste a Link turns any live website or web app into a narrated video. Paste the URL, walk through it once while narrating, and Velo’s browser agent takes care of the rest. It navigates your product and records the final video on its own, exactly as you scripted it.
Don’t have an asset to start from — just something you need to explain? Prompt to Video generates a video directly from a text prompt, no recording or upload required.