Skip to main content
Prompt to Video videos can be edited the same way they’re created: by describing the change you want in the same agent conversation.

​
Prompt-based editing

Open the video’s chat and ask for the change directly — for example:
  • “Trim the introduction down to two sentences.”
  • “Make the call-to-action at the end more direct.”
  • “Swap the second example for something about our enterprise plan.”
  • “Change the voice to [a specific cloned voice].”
The agent updates the script and re-renders the affected parts of the video. You don’t need to locate a specific scene or timestamp — describe the outcome and the agent handles the rest.

​
The scene editor is not available here

Prompt to Video videos are edited by prompt only. They open in their chat rather than in a project, so there is no route into the scene-by-scene editor — no timeline, no zoom or mask blocks, no per-scene timing. If you need that level of control, start from an asset instead. A screen recording, upload or link produces a project that opens in the full editor. See Creating a video.