Redesign logo

Redesign

Community
arjunkshah
redesign

Redesign existing interfaces to premium quality while preserving product goals and functional behavior.

Overview

Publisherarjunkshah
Repositorydesign-skill
Skill nameredesign
Stars
21
Forks
0
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by arjunkshah on GitHub. Read the source before you install it.

Installation

Install the Redesign AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/arjunkshah/design-skill/tree/main/skills/redesign
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/arjunkshah/design-skill.git /tmp/design-skill
mkdir -p .claude/skills
cp -r /tmp/design-skill/skills/redesign .claude/skills/redesign
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Redesign in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Redesign on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Redesign is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Redesign

Use this command when a product already exists and needs a stronger layout, visual system, and interaction polish.

Implementation Default

Default to Tailwind CSS for redesign output.

  • Use Tailwind utility classes for all visual updates unless the user explicitly requests another styling system.
  • Preserve existing framework/runtime, but move styling decisions toward Tailwind by default when feasible.

UI Library Access

Allowed reference libraries:

  • UIverse
  • FlyonUI (https://flyonui.com/)
  • daisyUI

Rules:

  • use libraries as accelerators, not as full-page templates
  • refactor and mutate library components to fit product semantics and brand direction
  • avoid repeated stock patterns from any single library
  • pair library usage with invented components when repetition risk is high

Pretext Integration

When redesigning existing UIs with text-overflow risk or unstable wrapping, use @chenglou/pretext to plan text layout before finalizing spacing and type scale.

  • Install when needed: npm install @chenglou/pretext
  • Use prepare() + layout() to validate key text blocks at target widths (hero, cards, CTA rows, nav labels).
  • Use the measurements to choose safer width constraints and reduce overflow regressions across breakpoints.
  • If the redesign depends on precise line rhythm, use prepareWithSegments() + walkLineRanges() to compare alternatives and pick the strongest composition.
  • If the project is static HTML/CDN-only or no JS build step exists, skip pretext integration and handle wrapping with responsive CSS constraints instead.
  • Do this silently; do not include user-facing "Note on Pretext" caveats unless asked.

Anti-Repetition Expansion (Required)

Redesign must not look like a cosmetic restyle of standard templates.

Rules:

  1. Build a broad candidate set across component and interaction types before choosing a direction.
  2. Replace repetitive card structures with differentiated component families (data cards, narrative blocks, comparison rows, proof strips, utility panels, etc.).
  3. Invent at least 2 new components for each substantial redesign when current UI blocks are generic.
  4. Invent at least 1 new animation pattern when motion is requested, tied to user feedback or hierarchy.
  5. Reject outputs where variation is only color, border radius, or font changes.

Redesign Diversity Quotas (Required)

For substantial redesigns, enforce:

  1. at least 6 materially different section structures across the page
  2. at least 3 non-card primary content modules
  3. at least 2 newly invented components adapted to this product context
  4. no single legacy pattern may dominate more than 25% of sections

If the redesign still resembles the original layout skeleton, escalate structural changes before finalizing.

Anti-AI-Look Redesign Gate (Required)

Reject the redesign if it reads as "template reskin".

Failure cases:

  1. mostly identical original layout with new colors/fonts only
  2. repeated stock cards replacing repeated stock cards
  3. unchanged section rhythm with minor surface edits
  4. motion limited to basic fade-ins without interaction logic

Minimum redesign bar:

  1. at least 3 clearly new structural modules replacing generic legacy blocks
  2. at least 2 bespoke components unique to the product context
  3. at least 1 unique interaction pattern not present in the source UI

Component Invention Rules

When inventing components for redesign:

  1. derive from product semantics (workflow, trust signal, conversion step), not visual novelty alone
  2. define explicit states and responsive behavior
  3. ensure implementation remains maintainable in Tailwind class composition
  4. add motion only where it improves comprehension or feedback speed

Core Objective

Transform a current interface into a clearer, higher-quality, more intentional design without breaking user flows.

Working Method

  1. Audit the current UI first:
  • identify hierarchy failures, spacing inconsistencies, weak composition, and generic patterns
  • detect interaction pain points and missing feedback states
  • list what must stay stable (flows, core IA, key copy, business constraints)
  1. Define a redesign direction:
  • choose a clear visual thesis and layout archetype
  • strengthen typography, section rhythm, and contrast
  • improve component consistency and responsive behavior
  1. Implement safely:
  • preserve functionality and navigation logic unless explicitly asked to change it
  • avoid regressions in forms, states, and accessibility
  • apply motion where it improves clarity and feedback, not decoration
  1. Validate quality:
  • verify desktop and mobile behavior
  • check reduced motion behavior if animations are added
  • confirm the redesign is materially better, not just stylistically different

Constraints

  • Do not ship generic “safe SaaS” defaults.
  • Do not remove critical information architecture without reason.
  • Do not trade usability for visual novelty.
  • Keep performance practical for the target context.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Redesign AI skill do?

Redesign existing interfaces to premium quality while preserving product goals and functional behavior.

Why use Redesign on TypingMind?

Because you install it once and use it with any model. Redesign is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Redesign in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/arjunkshah/design-skill/tree/main/skills/redesign. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Redesign?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Redesign?

As many as you like. As long as a model supports skills, you can use Redesign with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Redesign AI skill free?

Yes. It is published on GitHub by arjunkshah under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇