Free playground · no account

Same screen, two passes, two very different scores

Critique a deliberately flawed screen on gut feeling, then work a structured rubric over the exact same pixels. The gap between the two passes is your blind-spot count — measured, not argued.

Newsletter sign-up card — deliberately flawed mock
Join The Letter

One useful email a week. No noise.

your email address
name@example.com
GO!!MAYBE LATER

We store your address only to send the letter. Unsubscribe anytime.

Privacy policy
Pass 1 — vibes. What's wrong with this screen?
Your notes are matched by keywords, entirely in your browser — name concrete things you see.
Pass 2 — the rubric. Check every statement that is true of this screen.

Lock in your vibes pass first — that's the whole experiment.

The Demystify reveal

The same screen judged twice by the same person produces two different issue counts — the rubric literally lights up blind spots the freeform pass walked straight past.

What is a structured critique rubric?

A fixed set of questions you ask of every screen — hierarchy, contrast, affordance, consistency, feedback, clarity — instead of reacting to whatever catches your eye. It doesn't replace taste; it makes sure taste visits every room.

What are visual hierarchy and affordance?

Hierarchy is whether the most important thing looks the most important — the eye should land where the decision lives. Affordance is whether interactive things look interactive: a link that reads as plain text fails it, however pretty the screen.

Why does freeform feedback miss so much?

Unguided attention goes to the loudest problem and stops. A rubric forces a second, systematic sweep, which is why the same person finds more issues on the same screen minutes later — the sim counts that gap for you.

This playable is part of the registry: its tool page

Next: see if the model even fits

Can You Actually Run This? shows the GPU memory bill for any open model line by line — the same X-ray habit, applied to hardware.