Vertical explainers
An explainer is built for one frame, and that frame is picked when it's created:
| Frame | Shape | Made for |
|---|---|---|
| Landscape | Wide, 16:9 | Desktop screens and video |
| Portrait | Tall, 9:16 | Phones and reels |
| Square | 1:1 | Both at once, when you don't know which the viewer is on |
The frame is fixed once the explainer exists. Generated images, infographics and animations are all drawn for the shape you picked, so there's no switching a finished explainer from wide to tall. Slide text reflows to either layout, which is what lets a square explainer work on both.
A follow-up or a Go deeper question keeps the frame of the explainer it came from, and every lesson in a playlist shares one frame.
Picking a frame on Scrimba
In the composer on scrimba.com/explain, next to the privacy setting, a phone icon toggles Make for mobile. Leave it off and the explainer is built wide. Turn it on and it's built as a tall, portrait video, the shape of a reel.
Square isn't offered here, since you're on the device you're making it for. Explainers made by the Chrome extension and by hn.watch are always wide.
Picking a frame from a connected tool
The tool that opens an explainer for Claude Code, ChatGPT and every other MCP client takes a mode: square, portrait or landscape.
Leave it unset and the explainer is square. An agent has no way of knowing whether you'll watch on a phone or a desktop, so the safe default is the shape that works decently on both. Say which you want and the agent picks that instead:
The contract the tool returns carries the rules for the frame it opened with, such as top-to-bottom diagrams and shorter code lines in a portrait explainer. So the agent writes for the right shape from the first slide.
The playlist tool takes the same option and applies it to every lesson.
Watching
A portrait explainer opens in the tall layout everywhere, and a landscape one opens wide. A square explainer follows the device: tall on a phone held upright, wide on a desktop.
Vertical view in the player's settings menu overrides that for the explainer you're watching. It changes only your view, in this browser, and nothing about how the explainer was built.
Exporting and embedding
An exported portrait explainer is a 1440×2560 video. Every other frame exports at 2560×1440.
The embed code for a portrait explainer comes sized 450×800 instead of 800×450, and the embedded player runs in its tall layout.