---
name: moonshot-ux
description: "Elite polish pass: sweep a working-but-plain app until it feels built by a top-tier designer and elite PM — visual craft, subtle micro-interactions plus rationed 3D signature moments, habit-forming loops, and a usability floor, orchestrated in one pass. Use when the user says 'add polish', 'make it feel premium', 'make the experience unforgettable', 'it works but looks generic', 'elevate the whole app', or asks for creative animation and delight across the product. Do NOT use for a usability or WCAG audit of one flow, designing a brand-new screen or design system from scratch, a one-off animation on a single element, a maximum-creativity showpiece where distinctiveness is the whole brief, a retention or re-engagement strategy on its own, developer-facing surfaces (APIs, CLIs, docs), or runtime performance work."
license: MIT
---

# Moonshot UX

Elevate a functional app until it feels crafted. Run the whole pass autonomously — infer the product's voice from what already exists, state assumptions inline, never present a mode menu.

If the `hooked-ux` and `ux-heuristics` skills are installed, load them for the full frameworks; the essentials below are inlined so the pass stands alone without them.

## Invariants

- **Usability floor.** Polish that costs clarity is a regression. Self-evident labels, visible system status, recognition over recall, undo over confirmation, one primary action per view.
- **Rationed spectacle.** Subtle motion may appear wherever state changes; 3D or dramatic motion is limited to one or two **signature moments** (first run, core-action success, a milestone). Spectacle everywhere reads as noise.
- **Motion mechanics.** Animate transform and opacity only; micro-interactions run 150–300ms with ease-out; honor `prefers-reduced-motion` with a static path that loses nothing but the movement.
- **Ethics gate.** Habit mechanics ship only if the user is genuinely better off returning — no streak-shaming, no fake urgency, no dark patterns. If a mechanic depends on regret, cut it.
- **Anti-generic bar.** If a screen could have shipped from a default template, it isn't done. Name the committed direction — material, type voice, accent, density.

## The pass

1. **Inventory.** Walk every screen and every state — empty, loading, error, success. Mark the three moments that matter most: first run, the core action, the return visit.
2. **PM cut.** Give each screen one job; demote or delete before adding anything. Halve the words; buttons are outcome verbs; errors say what happened, why, and the fix.
3. **Craft.** Hierarchy, spacing rhythm, type scale, deliberate color — committed per the anti-generic bar.
4. **Motion.** Add micro-interactions that confirm state (press, save, navigate), then place the signature moments.
5. **Hook the loop.** Trigger → Action → Variable Reward → Investment: make the core action dead simple, vary the reward (social, discovery, mastery), and let each investment — content, data, streaks, preferences — load the next visit. Apply the ethics gate.
6. **Verify by using it.** Drive the running app in a real browser: complete the core action end to end, trigger the hover/press/focus states, watch every transition at full speed, and probe computed styles mid-flight where capture can't sample it. A screenshot cannot verify motion — it proves layout while freezing exactly what this pass adds — and reading the source is not seeing the render. Walk the keyboard path and the reduced-motion path. One fix round — one batch of fixes plus one re-verify — then stop, never a third pass; a fix of a few lines with no new states may still ride the re-verify. Anything left ships as a punch list, each item naming why it didn't ship: blocked by architecture, needs a product decision, or out of scope. "More polish" is not a reason — do it or drop it. Report the committed direction alongside the punch list; a follow-up invocation starts the next pass from there.

Load [`references/polish-playbook.md`](./references/polish-playbook.md) for per-step depth: signature-moment candidates, when 3D earns its cost, hook-loop wiring by product type, and copy elevation patterns.