Explainer Toy
A small interactive widget inside an explanation that lets the reader change one thing and watch what happens. A toggle that switches a bug on and off, or a button that makes two people edit at the same time, are toys. Each one is built around a single idea, runs on tiny made-up data, and exists to be played with, not used. The word comes from *toy example* and *toy model*, where "toy" means a deliberately simplified stand-in for the real thing. It is an informal term with no official definition.
Notes
I ran into this word when Claude explained a hard design decision to me as a web page with diagrams and two clickable demos, and kept calling the demos "toys". I liked it enough to ask where it came from.
What I checked: Anthropic's published system prompt for the model (September 28, 2026) does not contain the word, and neither did the artifact design guidance loaded in that session or the skill files on the machine. Claude's own guess was that it chose the word itself, from "toy example" and "toy model". It said it can't verify that, so treat the origin as unknown.
What made the toys useful: each one changes exactly one thing and shows exactly one result. One flips how a server labels a bundle of edits, and the page goes from "Hi!" to "Hii!!". The other has two people erase different lines of a quote box at the same time, and shows what two different designs do with it.
One caveat. A toy can make a guess look like a demonstration. The second toy on that page shows a result that was reasoned from code and a maintainer's explanation, not reproduced, and the page said so in small print. A reader who only plays with the toy never sees the small print. If I build one, the uncertainty belongs next to the button. [JOEL: add your own example of a toy that helped you, or one that misled you, or cut this line.]
Related: toy example, toy model, explorable explanation (a term I know of but haven't traced), and demo (usually bigger and meant to impress).