design-engineering
Craft rules for web UI polish (press states, origins, transforms, clip-path, gestures, performance, accessibility) and a Before/After review table.
Skill by luizhcrocha
Chalk Labs
Need help deploying design-engineering?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/luizhcrocha-design-engineering
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/design-engineering
What's inside
- Skill name
design-engineering- Description
- Craft rules for web UI polish (press states, origins, transforms, clip-path, gestures, performance, accessibility) and a Before/After review table. Use to polish UI or review it for feel.
- License
- MIT
- Instructions
- 46 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- craft.md · references/
SKILL.md (first lines)
# Design engineering Good interfaces are made of many small details that users never notice one by one. Each detail below has a reason. Apply it where the reason applies, and study why the interfaces you admire feel the way they do. ## Pick the skill for the job - One animation to build: `animate`. - A diff whose motion needs a verdict: `review-animations`. - A codebase's motion to audit and plan: `improve-animations`. Places that should start moving: `find-animation-opportunities`. - Gestures, springs, materials and type at Apple's level: `apple-design`. - A web app that feels wrong on a phone: `mobile-native`. - A component library to choose: `pick-ui-library`. - Several directions for one piece of UI: `prototype`. This skill covers the rest: component polish, and a UI review that looks at more than motion. ## Rules that catch most problems 1. Decide whether it should animate at all. An action seen 100 or more times a day, or started from the keyboard, gets no animation. 2. Name exact properties. Never `transition: all`. 3. Enter and exit with `ease-out` or a strong custom curve, such as `cubic-bezier(0.23, 1, 0.32, 1)`. Never `ease-in` on UI. 4. Keep UI animation under 300 ms. A dropdown takes 150 to 250 ms; press feedback takes 100 to 160 ms. 5. Give every pressable element an `:active` state, such as `transform: scale(0.97)`. 6. Never enter from `scale(0)`. Start from `scale(0.95)` with `opacity: 0`. 7. Popovers scale from their trigger (`transform-origin: var(--transform-origin)` in Base UI). Modals stay centered. 8. Use transitions, not keyframes, for anything a user can trigger again mid-flight. They retarget from the current value. 9. Animate `transform` and `opacity` only. In Motion, use the full `transform` string instead of the `x` and `y` shorthands. 10. Honor `prefers-reduced-motion` with gentler motion, not none, and put hover effects inside `@media (hover: hover) and (pointer: fine)`. When feel cannot be judged from code, say so. Play the animation at 2 to 5 times its duration or frame by frame in DevTools, test gestures on a real device, and look again the next day. ## Review format A UI review is one markdown table with the columns Before, After and Why, one row per issue. Never write "Before:" and "After:" on separate lines. | Before | After | Why | | --- | --- | --- | | `transition: all 300ms` | `transition: transform 200ms ease-out` | Name the properties; `all` animates what you did not mean to | | `transform: scale(0)` | `transform: scale(0.95); opacity: 0` | Nothing in the real world appears from nothing | | No `:active` state on a button | `transform: scale(0.97)` on `:active` | A press needs instant feedback | ## The reference Read [references/craft.md](references/craft.md) for the reasons and the code behind each rule: the animation decision framework, springs, component principles (tooltips, blur over a crossfade, `@starting-style`), transforms, clip-path patterns (tabs, hold to delete, image reveals, comparison sliders), drag and gesture handling, performance, accessibility, principles for components people love, stagger, debugging, and the review checklist.
From luizhcrocha/skills, MIT.
More from luizhcrocha/skills
- 0.570score
lang-c
luizhcrocha
C rules and taste for tstack (C23, GCC/Clang, sanitizers, clang-tidy, QMK keymaps).
SkillAFreshness: AClaude Code +9 - 0.567score
architect
luizhcrocha
Sketch types, signatures and the module map from competing candidates before code, then implement against it.
SkillAFreshness: AClaude Code +9 - 0.567score
apple-design
luizhcrocha
Apple's fluid-interface principles for the web: springs, velocity handoff, momentum, rubber-banding, materials, type.
SkillAFreshness: AClaude Code +9 - 0.567score
create-verification-skill
luizhcrocha
Generate a project-local verify-<app> skill that launches the app and drives it like a user, with a feature map.
SkillAFreshness: AClaude Code +9 - 0.567score
how
luizhcrocha
Explain how code works, as onboarding a senior engineer.
SkillAFreshness: AClaude Code +9 - 0.567score
interrogate
luizhcrocha
Adversarial review: three fresh reviewers (correctness and security, maintainability, intent) try to break a change; the lead gives a verdict.
SkillAFreshness: AClaude Code +9 - 0.561score
explain
luizhcrocha
Explain a piece of work plainly (what, how, why) as one conversational account.
SkillAFreshness: AClaude Code +9 - 0.561score
figure-it-out
luizhcrocha
Design an auditable bespoke playbook when no tuca-mode playbook fits.
SkillAFreshness: AClaude Code +9 - 0.561score
find-animation-opportunities
luizhcrocha
Find places in a UI that don't animate but should, and reject the ones that shouldn't; read-only, exact values.
SkillAFreshness: AClaude Code +9 - 0.561score
improve-animations
luizhcrocha
Audit a codebase's animation and motion code and write prioritized, self-contained fix plans for other agents; read-only on source.
SkillAFreshness: AClaude Code +9 - 0.561score
arena
luizhcrocha
Run N parallel attempts at one task under distinct constraints, pick a base, graft the best of the others.
SkillAFreshness: AClaude Code +9 - 0.561score
animate
luizhcrocha
Build an animation or transition that passes review-animations, or name a motion effect ("what's it called when...").
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 1, 2026 (29 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/luizhcrocha/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from luizhcrocha
- 0.570score
lang-c
luizhcrocha
C rules and taste for tstack (C23, GCC/Clang, sanitizers, clang-tidy, QMK keymaps).
SkillAFreshness: AClaude Code +9 - 0.567score
how
luizhcrocha
Explain how code works, as onboarding a senior engineer.
SkillAFreshness: AClaude Code +9 - 0.567score
apple-design
luizhcrocha
Apple's fluid-interface principles for the web: springs, velocity handoff, momentum, rubber-banding, materials, type.
SkillAFreshness: AClaude Code +9 - 0.567score
architect
luizhcrocha
Sketch types, signatures and the module map from competing candidates before code, then implement against it.
SkillAFreshness: AClaude Code +9 - 0.567score
create-verification-skill
luizhcrocha
Generate a project-local verify-<app> skill that launches the app and drives it like a user, with a feature map.
SkillAFreshness: AClaude Code +9 - 0.567score
interrogate
luizhcrocha
Adversarial review: three fresh reviewers (correctness and security, maintainability, intent) try to break a change; the lead gives a verdict.
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get design-engineering running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
design-engineering skill. Agent Mart. https://agentmart.dev/skills/luizhcrocha-design-engineering (last updated 2026-10-24)