Dryui Build logo

Dryui Build

Community
rob-balfre
dryui-build

Experimental DryUI layout-only skill that extracts responsive page structure from design images as abstract colored blocks with strict shell areas, fixed page container queries, and visual self-checks. Use when testing design-image to layout-skeleton extraction.

Overview

Publisherrob-balfre
Repositorydryui
Skill namedryui-build
Stars
9
Forks
0
Bundled files
Instructions only
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by rob-balfre on GitHub. Read the source before you install it.

Installation

Install the Dryui Build 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/rob-balfre/dryui/tree/main/reports/skill-experiments/2026-05-11/batch-7/v10-consolidated-layout
  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/rob-balfre/dryui.git /tmp/dryui
mkdir -p .claude/skills
cp -r /tmp/dryui/reports/skill-experiments/2026-05-11/batch-7/v10-consolidated-layout .claude/skills/dryui-build
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Dryui Build 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 Dryui Build 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 Dryui Build 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.

DryUI Layout Blocks: Consolidated

Build only an abstract colored block layout skeleton. This is a layout extraction stage, not a UI implementation.

Only edit:

  • src/routes/+page.svelte
  • src/layout.css

Do not import anything. Do not use DryUI components, Lucide, icons, SVGs, images, charts, tables, forms, buttons, inputs, avatars, realistic widgets, copied text, copied metrics, copied dates, copied names, or generated business data.

1. Classify The Task

  • Full page: build a page shell with topbar/navigation, primary, and optional secondary, utility, rail, summary, actions.
  • Page section: build a section shell with primary and optional adjacent regions.

For full-page design-image references, treat the whole image as the page shell.

2. Blueprint Before Coding

Write the region plan mentally before editing:

  • Mobile: one-column top-to-bottom order.
  • Tablet: first useful two-pane/grouped composition.
  • Desktop: persistent rail/sidebar, right panels, attached panes, dominant primary surface.

Decide:

  • Which region is primary.
  • Whether spacing is contiguous, guttered, or mixed.
  • Which regions are secondary or utility, so they do not become larger than primary.
  • Which visible bands are shell regions versus nested primitives.

3. Use Recipes Only For Proportions

Do not change required area names for a recipe.

  • Dashboard: summary strip, primary chart/workspace, insights/activity.
  • Board/Kanban: board is primary; task detail is secondary; activity is utility; desktop rail if present.
  • CRM: table/pipeline is primary; account/detail is secondary; forecast/activity are utility.
  • Settings/Admin: settings/profile/form stack is primary; help/status/activity are secondary/utility; compact actions; desktop rail if present.
  • Knowledge/Document Reader: reader is primary; collections/list are navigation/secondary; AI summary is secondary/utility; reader must be largest at desktop.

4. Markup Contract

For a full page, use this direct-child shape:

svelte
<main data-layout="<name>-shell">
	<div data-layout-area="page">
		<header data-layout-area="topbar">Topbar</header>
		<nav data-layout-area="navigation">Navigation</nav>
		<section data-layout-area="primary">Primary</section>
		<aside data-layout-area="secondary">Secondary</aside>
	</div>
</main>

Rules:

  • page, primary, and at least one of topbar or navigation are mandatory.
  • Direct-child shell areas may only be: rail, topbar, navigation, summary, primary, secondary, utility, actions.
  • Each shell area should appear once.
  • data-layout-area is shell-only. Use it only on [data-layout-area='page'] and its direct children.
  • Never put data-layout-area on nested spans, nested divs, articles, list items, labels, chips, cards, search placeholders, tabs, badges, rows, columns, or decorative blocks.
  • Nested blocks must use data-layout="<specific-name>" and classes, never data-layout-area.
  • Put recipe-specific meaning in text labels/classes, not direct-child area names.

5. Abstract Block Content

The skeleton should look like colored regions and blocks, not a finished UI.

  • Visible text is restricted to these region labels only: Topbar, Navigation, Rail, Summary, Primary, Secondary, Utility, Actions.
  • Use at most one visible label per shell region.
  • Nested primitives should usually be empty elements.
  • Do not copy visible text from the reference image.
  • Do not create domain labels such as customer names, project names, document titles, settings names, money, dates, percentages, counts, or status text.
  • Represent cards, controls, rows, tabs, charts, documents, columns, and copy as colored rectangles, lines, chips, and blocks.
  • Use repeated abstract primitives to communicate density and grouping.
  • Use distinct color families for region types, but keep the layout readable.

6. CSS Contract: Copy This Shape

Use the fixed container name page. Do not invent another container name.

css
[data-layout='<name>-shell'] {
	container: page / inline-size;
}

[data-layout='<name>-shell'] [data-layout-area='page'] {
	display: grid;
	grid-template-areas:
		'topbar'
		'navigation'
		'primary'
		'secondary';
	grid-template-columns: minmax(0, 1fr);
}

[data-layout='<name>-shell'] [data-layout-area='page'] > [data-layout-area='topbar'] {
	grid-area: topbar;
}

[data-layout='<name>-shell'] [data-layout-area='page'] > [data-layout-area='primary'] {
	grid-area: primary;
}

@container page (min-width: 48rem) {
}

@container page (min-width: 72rem) {
}

Hard rules:

  • container: page / inline-size must appear exactly once on the shell selector.
  • Use exactly @container page (min-width: 48rem).
  • Use exactly @container page (min-width: 72rem).
  • Never use @container <name>, @container (width...), or @container page (width>=...).
  • Any CSS selector that targets a shell area other than page must use [data-layout-area='page'] > [data-layout-area='<area>'].
  • Do not use broad descendant selectors like [data-layout='<name>-shell'] [data-layout-area='primary']; they can accidentally style nested blocks.

7. Layout Rules

  • Mobile is one column only.
  • Tablet must visibly use extra width when the reference has grouped panes.
  • Desktop must preserve the reference shell: rail/sidebar, right inspector, attached panes, and dominant primary surface.
  • Put all display: grid, display: flex, grid, flex, container, and breakpoint rules in src/layout.css.
  • Use minmax(0, 1fr) for flexible tracks.
  • Use only DryUI token gaps such as var(--dry-space-2), var(--dry-space-3), var(--dry-space-4), and var(--dry-space-6).
  • If panes touch in the reference, use gap: 0; if gutters are visible, use token gaps.
  • primary must be the largest or most visually dominant desktop content region.
  • Utility/status/activity blocks must never be larger than primary.
  • Do not create large blank shell regions. Every direct shell area should contain visible blocks near its top edge.
  • Do not let mobile output become a long dump of every desktop primitive; summarize repeated content when the reference collapses.

8. Route Styling Rules

Route <style> is visual only:

  • color, background, border, shadow, typography, padding inside blocks, labels, and min-block-size are allowed.
  • display, grid-*, flex-*, container, @container, and @media are not allowed in route <style>.
  • Keep labels short and generic.
  • Use distinct accessible colors.

9. Visual Self-Check Before Finishing

After bun run build, inspect or reason through the 390px, 820px, and 1440px views:

  • Shell regions change shape at the two container breakpoints.
  • The primary region is visibly dominant on desktop.
  • Navigation/rail and right-side utility panels attach in the same broad positions as the reference.
  • No large blank region appears before the real content.
  • No nested primitive is accidentally sized like a shell area.
  • Text does not wrap awkwardly; if a label is unnecessary, remove it.

10. Source Self-Check Before Finishing

Run bun run build, then check:

  • No import in src/routes/+page.svelte.
  • No @dryui/ui, lucide, <button>, <input>, <form>, <table>, <img>, or <svg>.
  • No copied text, numbers, dates, customer names, document titles, metrics, or statuses from the reference.
  • No @media.
  • No display, grid-*, flex-*, or @container in route <style>.
  • src/layout.css contains exact container: page / inline-size.
  • src/layout.css contains exact @container page (min-width: 48rem) and @container page (min-width: 72rem).
  • Full page has exactly one direct child data-layout-area="primary".
  • No nested primitive uses data-layout-area.
  • 390px, 820px, and 1440px should not overflow horizontally or show a large blank shell region.

Frequently asked questions

What does the Dryui Build AI skill do?

Experimental DryUI layout-only skill that extracts responsive page structure from design images as abstract colored blocks with strict shell areas, fixed page container queries, and visual self-checks. Use when testing design-image to layout-skeleton extraction.

Why use Dryui Build on TypingMind?

Because you install it once and use it with any model. Dryui Build 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 Dryui Build in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/rob-balfre/dryui/tree/main/reports/skill-experiments/2026-05-11/batch-7/v10-consolidated-layout. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Dryui Build?

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 Dryui Build?

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

Is the Dryui Build AI skill free?

Yes. It is published on GitHub by rob-balfre 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 👇