Skip to main content

How do I use a video with the Heptabase AI agent?

Before you start

  1. Add your video to the card library to create a video card.

  2. Generate a transcript for that video card.

The AI uses the transcript as context (it doesn’t read the raw video).

​

For YouTube videos, Heptabase automatically selects a caption track when generating the transcript. You can’t manually choose the YouTube caption language.

​

Start a chat with a video card

  • From a whiteboard: Right-click the video card and choose New chat / Add to chat (top option).

  • From the chat panel: Click the + button at the bottom-left and select the video card you want, or @mention that video card directly.

​

Not supported yet

  • You can’t insert or embed a standalone video in a chat and have the AI read it as context.

Did this answer your question?