agentleFS
coolxeocoolxeo/grok-bot-skills/skills/design/redesign-existing-projects/SKILL.md

redesign-existing-projects

Skill8 starsChanged 2 days agoMITView on GitHub

Use when improving the design of a site or app that already exists. Audits the live pages, ranks the problems by impact, and fixes them in small reviewable steps on the current stack without breaking features.

104 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use stitch-design-taste, design-taste-frontend-v1 and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • HK-hubSkill

    stitch-design-taste

    Semantic Design System Skill for Google Stitch. Generates agent-friendly DESIGN.md files that enforce premium, anti-generic UI standards — strict typography, calibrated color, asymmetric layouts, perpetual micro-motion, and hardware-accelerated performance.

    Also in 51 projects that use this

  • HK-hubSkill

    design-taste-frontend-v1

    The original v1 taste-skill, preserved for projects depending on its exact behavior. The current default is `design-taste-frontend` (v2 experimental), which is a substantial rewrite. Use this v1 install name only if you need exact backward compatibility.

    Also in 29 projects that use this

  • hjdspaceSkill

    flutter-setup-declarative-routing

    Configure `MaterialApp.router` using a package like `go_router` for advanced URL-based navigation. Use when developing web applications or mobile apps that require specific deep linking and browser history support.

    Also in 3 projects that use this

  • hjdspaceSkill

    flutter-setup-localization

    Add `flutter_localizations` and `intl` dependencies, enable "generate true" in `pubspec.yaml`, and create an `l10n.yaml` configuration file. Use when initializing localization support for a new Flutter project.

    Also in 3 projects that use this

  • sanmidableSkill

    animate-expo

    Also in 5 projects that use this

  • sanmidableSkill

    ask-sonner

    Also in 5 projects that use this

  • +5 more from hjdspace

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 100 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask, or infer and state: what is the page supposed to make people do, and what must not change (brand colors, logo, a flow customers know)?

    line 14
  2. Distinctive:

    Read the code: framework, styling method and version (Tailwind 3 or 4, CSS modules, plain CSS), component library, and design tokens if any.

    line 15
  3. Distinctive:

    Layout and spacing: no grid, uneven padding, inconsistent max-widths, things that don't align.

    line 25
  4. Distinctive:

    Run the app and its tests after each step.

    line 41
  5. Distinctive:

    Before and after screenshots as the deliverable.

    line 56

“Common” is measured: kept by at least 6 in 10 of the 100 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Redesign existing projects

Make an existing site look and work better without starting over. Keep the stack, the features, and what users already rely on.

1. Before touching code

  • Ask, or infer and state: what is the page supposed to make people do, and what must not change (brand colors, logo, a flow customers know)?
  • Read the code: framework, styling method and version (Tailwind 3 or 4, CSS modules, plain CSS), component library, and design tokens if any.
  • Screenshot the key pages at mobile and desktop widths. These are the "before".

2. Audit

Walk each page and note problems with a screenshot or line reference:

  • Hierarchy: is the main action the most visible thing? Can you tell what the page is in five seconds?
  • Type: too many sizes or families, weak scale, lines too long, poor contrast.
  • Color: clashing or oversaturated colors, accent used everywhere, gray text that fails contrast.
  • Layout and spacing: no grid, uneven padding, inconsistent max-widths, things that don't align.
  • States: missing hover, focus, active, loading, empty, and error states.
  • Feedback and control: users can't tell what happened after a click, can't undo, or get stuck.
  • Accessibility: keyboard access, visible focus, alt text, labels on inputs, WCAG AA contrast, reduced motion.
  • Template tells: identical cards everywhere, generic hero, stock testimonials, filler copy (run copy through no-ai-slop).
  • Code: one-off magic numbers, !important piles, transition: all, duplicate components.

3. Rank

Score each problem by user impact and effort, then fix in that order. Problems that block the main action or fail accessibility come first, whatever their size. Share the ranked list before starting if the changes are large.

4. Fix in small steps

  • One theme per commit or PR (for example "type scale", "button states"), so each can be reviewed and reverted alone.
  • Work inside the current stack. No framework migrations, no new UI library unless the user agrees. Check the dependency file before importing anything.
  • Move repeated values into tokens or variables as you go.
  • Run the app and its tests after each step. Nothing that worked before may break.

5. Show the difference

Take "after" screenshots at the same widths and pair them with the "before" set. List what changed, what you deliberately left alone, and anything that needs the user's decision (new imagery, copy, brand changes).

Inspired by

  • Leon Lin, Taste Skill redesign-skill (MIT). Does well: scan, diagnose, fix on the existing stack, with a list of what generated designs usually forget. Missing: no before and after screenshots, almost nothing on accessibility, and a fixed priority list that opens with a font swap whatever the goal.
  • Nielsen Norman Group, 10 Usability Heuristics. Does well: system status, user control, consistency, and error prevention as a lens for auditing. Missing: no visual design or code guidance.
  • Vercel, Web Interface Guidelines (MIT). Does well: concrete checks for focus, keyboard, forms, motion, and layout. Missing: a checklist, with no process for ranking or rolling out fixes.

What this version adds

  • Goal and "must not change" questions before the audit.
  • Before and after screenshots as the deliverable.
  • Impact-by-effort ranking instead of a fixed order, with accessibility treated as a blocker.
  • One theme per commit so each change can be reviewed and reverted.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.