train-aesthetic-eye — quality + safety report

---
name: train-aesthetic-eye
description: Force the user to stop hand-waving about "taste" and consciously train their aesthetic eye on any design, UX, or polish decision. Use this skill aggressively whenever the user is doing visual design review, product polish, brand work, UI critique, picking between mockups, debating typography or spacing, evaluating "is this good enough to ship", asking "does this look right", "I can't tell why I don't like it", "it feels janky but I can't say why", "should we ship this MVP-quality", "users won't notice", "the details don't matter at this stage", "we'll polish later", "I'm not a designer so I can't judge", "good enough for v1", "the engineer says it's fine", or any moment a team is about to wave through a mediocre artifact because nobody on the call can articulate what's wrong. Also fires on icon sets, marketing pages, slide decks, hardware industrial design, packaging, and any product surface a customer will see. Trigger eagerly even when the user does not name Musk or the framework.
---

# Train Aesthetic Eye

> "You can train yourself. You can make yourself pay attention to 'why.' You can learn to bring subconscious awareness into conscious awareness. Look closely and carefully. Look at each object's geometry."
> — Elon Musk, *The Book of Elon* (Chapter: A Whole New Kind of Car Company (aesthetics subsection))

## What this skill captures

Taste is not a gift. It is a trained sensor. Musk's claim is concrete: "Our brains have some intrinsic elements that represent beauty, which trigger the emotion of appreciation of beauty in our mind." Most people register the summation as "that's ugly" or "that's beautiful" but cannot break it down. The fix is conscious decomposition — drag the subconscious verdict into conscious words, name the geometry, name the proportions, name the detail that is off. Then fix it.

The value: you stop shipping mediocre surfaces because no one on the team could articulate what was wrong. You learn to see, and once you see, you cannot un-see — which is exactly the standard a perfect product requires.

## When to use this skill

- A team is about to approve a mockup, marketing page, or product surface and the loudest signal in the room is "looks fine to me."
- The user says something feels off but cannot name what.
- A non-designer is making a visual call and defaulting to engineer-aesthetics ("it works, ship it").
- An MVP is being waved through with "we'll polish later" — a phrase that almost always means never.
- Industrial design, packaging, or hardware decisions where small geometry choices compound across millions of units.
- Internal-tool or B2B UI where the team has convinced itself users do not care.

## The how-to

1. **Name the subconscious verdict out loud, before any analysis.**
   > "Most of us experience this as 'that's ugly,' or 'that's beautiful,' or 'wow, that's elegant,' but can't break down why."
   > — *The Book of Elon*
   First impression is signal, not noise. Capture it in one word before reasoning kicks in and rationalizes it away.

2. **Decompose the impression into geometry.**
   > "Look closely and carefully. Look at each object's geometry."
   > — *The Book of Elon*
   For each element, name the proportion, alignment, spacing, curve, and weight. If you cannot describe the geometry in words, you have not actually looked.

3. **Enumerate the small details.**
   > "Pay attention to the little details. Train yourself to notice them. Notice the nuances of design, shape, form, function, and the way it looks in different lights."
   > — *The Book of Elon*
   List at least ten specific details — corner radius, hover state, kerning, icon weight, shadow softness, contrast at the edge. The summation of details is what the user feels.

4. **Test the aesthetic-utility tradeoff explicitly.**
   > "What was hard about the Model S and Model X was to combine aesthetics and utility, to balance the two."
   > — *The Book of Elon*
   Beauty bought by sacrificing function is decoration. Function shipped without beauty is contempt for the user. Name where on the tradeoff curve you are sitting and whether you can push both at once.

5. **Check whether it feels bigger on the inside than it looks on the outside.**
   > "Another incredibly important design principle is to have it feel bigger on the inside than it looks on the outside. That's also a hard thing to do."
   > — *The Book of Elon*
   In software: does the surface feel light but reveal depth on use? In hardware: compact silhouette, generous interior. Apply this whether you are designing a sedan, a settings panel, or a CLI.

6. **Accept the side effect: you will now be bothered forever.**
   > "Anyone can do this, although it is a double-edged sword, because then you always notice all the little things. Now when something's off—even a little thing—it drives me bananas."
   > — *The Book of Elon*
   This is the price of admission. If small things stop bothering you, your eye has decayed. Welcome them back.

7. **Set the bar at perfect, not acceptable.**
   > "If you're trying to make a perfect product, attention to detail is essential."
   > — *The Book of Elon*
   "Good enough" is a moving floor that always trends down. Replace it with a written list of specific details that must be right before ship.

## Common failure modes

- **"Users won't notice."** Musk's answer: "Most people don't consciously notice the small details, but they do subconsciously. Your mind takes in an overall impression." They notice. They just cannot tell you.
- **Deferring taste to "the designer."** The whole team ships the product. If only one person on the team can see, every other surface — error states, empty states, emails, install flows — will be ugly.
- **Confusing personal preference with broken aesthetics.** Musk: "Not everyone likes exactly the same thing, but there's a lot of commonality." Some things are objectively misaligned, miskerned, or misproportioned. Do not retreat to "taste is subjective" to avoid the work.
- **Polishing the easy surface and skipping the hard one.** "To make a sports car look good is relatively easy. But to make a sedan or an SUV look good is quite difficult." The settings page, the admin panel, the 404 — those are the SUV.
- **"We'll polish later."** Later does not come. The same team that shipped the rough version is the team that will ship the next rough version on top of it.

## When NOT to use this skill

- Pre-product-market-fit prototypes where the only question is "does anyone want this at all" — aesthetics is the wrong filter, talk to users instead.
- Internal scripts, dev tools, or one-off automations that no human will look at twice.
- Backend architecture, data schemas, or other surfaces with no perceptual layer for a human user.
- When the actual blocker is a functional bug, not a visual one — fix the bug first; do not aestheticize a broken thing.

## Source

The Book of Elon by Eric Jorgenson (2026, Scribe Media). Chapter: "A Whole New Kind of Car Company (aesthetics subsection)" (in "Building Tesla").