Skip to main content
After clicking Go, Make my Velo → on the Script Review page, Velo launches an AI agent to record your Velo. The agent navigates to your URL and works through the script autonomously - no input needed from you.

​
What you’ll see

The Agent Recording screen is split into two panels:
  • Agent Activity (left panel) - a live log of what the agent is doing, step by step. You’ll see actions like “Navigated to [your URL]”, “Waited”, “Identifying next interaction”, and so on. This updates in real time as the agent works through each scene.
  • Browser Agent (right panel) - a live feed of the agent’s browser session. You can watch the agent interact with your URL exactly as a real presenter would - scrolling, clicking, and navigating through the page according to the script.
The agent runs in an isolated browser environment. It navigates to your URL independently - this is separate from any tab you may have open.

​
Waiting for the agent

The agent works through the full script automatically. Depending on the number of scenes and target duration, this typically takes 1–3 minutes. You can watch the live feed or leave the page - Velo processes everything in the background and will be ready when the agent finishes.

​
Cancelling the session

If you need to stop the agent, click Cancel Session in the top right. The recording will stop and you’ll see a confirmation screen:
“Recording was stopped - The agent recording was cancelled. You can start again to generate your video.”
Click Go make my Velo to go back and run the agent again from the Script Review page.
Cancelling the session will not consume Velo Minutes for the cancelled run.

​
After the agent finishes

When all scenes are complete, Velo processes the recording. You’ll see a brief Processing… state. Once done, you’ll be taken to the preview page where you can watch your generated Velo and open it in the editor to refine it.