Forge logo

Forge

Organization
Webba-Creative-Technologies
forge

Forge design system for React. Use when writing any React UI in this project, when picking which component to use, when theming, when creating block templates, or when reviewing UI code for accessibility and anti-patterns. Covers components, tokens, theming, motion, patterns, a11y, and anti-patterns.

Overview

PublisherWebba-Creative-Technologies
Repositoryforge
Skill nameforge
Stars
19
Forks
0
Bundled files
11
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.

  • 11 bundled files

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

  • Open source

    Published by Webba-Creative-Technologies on GitHub. Read the source before you install it.

Installation

Install the Forge 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/Webba-Creative-Technologies/forge/tree/main/skills/forge
  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/Webba-Creative-Technologies/forge.git /tmp/forge
mkdir -p .claude/skills
cp -r /tmp/forge/skills/forge .claude/skills/forge
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Forge 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 Forge 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 Forge 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.

Forge Design System

Forge is a React component library that ships as wss3-forge. Every UI concern (layout, forms, overlays, data display, charts, motion) is covered by a component. All colors, spacing, radius, shadows, and motion live in CSS variables injected by ForgeProvider.

Decision rule: if the task changes what the user sees, touches, or waits for, load this skill.

Find by task

Lookup table for the most common requests. Jump straight to the relevant anchor.

If you're trying toReadSkip ahead
Build a login screenpatterns.md#auth
Build signup or forgot-passwordpatterns.md#auth-forgot
Show a toast on user actioncomponents.md#feedback-toast
Inline alert (success, error, info)components.md#feedback-decision
Errored input with screen-reader feedbacka11y.md#inputs
Icon-only buttona11y.md#buttons
Pricing strip with monthly/yearly togglepatterns.md#pricing
Pick Modal vs Sheet vs Drawer vs Popovercomponents.md#overlay-decision
Pick Card vs StatCard vs KpiCardcomponents.md#cards-decision
Pick VStack vs HStack vs Grid vs Boxcomponents.md#layout-decision
Pick Input vs Select vs Comboboxcomponents.md#forms-decision
Pick Checkbox vs Switch vs Radio vs Tabscomponents.md#selection-decision
Animate items into view, one by onemotion.md#stagger
Sticky scroll-jacked phasemotion.md#sticky-section
Brand color customizationtheming.md#brand-customization-in-30-seconds
Theme color won't applyfaq.mdfaq.md
Empty statepatterns.md#empty-state
Loading or error or empty triadpatterns.md#triad
Settings page (profile, billing, team)patterns.md#settings
Marketing page (hero, pricing, testimonials)patterns.md#marketing
Data screen (table, master-detail, kanban)patterns.md#data
Dashboard with sidebarpatterns.md#dashboard-sidebar--content
Build a docs sitepatterns.md#docs
Pick Navbar vs AppSidebar (horizontal vs vertical menu)components.md#navigation

Review priority (check in this order)

  1. Accessibility. Focus visible, correct roles, aria-label on icon-only buttons, aria-invalid/aria-describedby on errored inputs, keyboard navigation works, reduced motion respected. See a11y.md.
  2. Semantics. Using the right Forge primitive: Heading not styled <h1>, Button not styled <button>, VStack/HStack not flex <div>, Card not styled container. See anti-patterns.md.
  3. Motion. Durations 150 to 300ms for interactions, reduced-motion respected by ForgeProvider, no animations over 500ms for UI feedback. See motion.md.
  4. Theme parity. Code works in both darkTheme and lightTheme. No hardcoded rgba(255, 255, 255, ...) or #xxx. Use CSS vars or color-mix(in srgb, currentColor N%, transparent).
  5. Performance. VirtualList when rendering more than 100 rows, lazy routes, <Motion whileInView> over animate for offscreen content.

Hard rules

  1. Import only from 'wss3-forge'. No deep imports.
  2. Wrap the app in <ForgeProvider>. Without it every component renders unstyled.
  3. Icons only from @fluentui/react-icons. Convention: {Name}{Size}{Style} (e.g. ArrowRight20Regular).
  4. Colors via var(--...). No hardcoded hex, RGB, or color names.
  5. Spacing via gap/padding props on Forge primitives. No raw px/rem when a semantic key exists.
  6. Layout via VStack, HStack, Stack, Grid, Flex, Box, Center, Spacer, AspectRatio. No flex divs.
  7. Responsive via useIsMobile, useBreakpoint, or responsive object props. No @media queries in app code.
  8. Typography via Heading, Text, Label. Never style <h1>, <span>, or <p> manually.
  9. No em-dashes anywhere. Period, comma, colon, or line break.
  10. No emojis in UI. Use Fluent icons.
  11. No decorative colored backgrounds behind icons. Color the glyph itself via style={{ color: '...' }}. Only StatusBadge, Avatar status, and colored Badge variants keep a tinted surface.
  12. No custom components when Forge provides one. Check components.md first.
  13. No custom CSS classes for design-system concerns (spacing, color, radius, shadow, typography). Use props or CSS vars.

For the reasoning behind each rule, see design.md. For concrete wrong/right pairs, see anti-patterns.md.

What Forge is NOT

Forge is a restrained design system. These are baseline anti-aesthetics:

  • No aurora-gradient backgrounds. The page reads as flat surfaces with brand-tinted accents, not as a colored sky.
  • No glass-on-everything. Backdrop-filter is a tool for the navbar and the modal scrim, not for every card.
  • No icon glow / icon-tint backgrounds. Color the glyph itself; let surfaces stay calm.
  • No cyan-magenta-violet gradients. They read as AI-template defaults; we build for Linear / Stripe / Vercel restraint.
  • No streaming-text fakery, no ASCII art in production UI, no AI-marketing clichés.

When in doubt, restrain. Forge ships polished defaults; consumers reach for them, not around them.

Pre-delivery checklist

Before claiming done, verify:

  • tsc --noEmit clean.
  • Works in both dark and light mode (no hardcoded dark-only colors).
  • Focus ring visible on every interactive element.
  • Icon-only buttons have aria-label or tooltip.
  • Touch targets at least size="sm" (32px) on mobile flows.
  • Motion durations within 150 to 300ms for interactions, reduced-motion respected.
  • No em-dash, no emoji in UI strings.
  • No hardcoded hex/rgba in styles. All colors via CSS var.
  • No fallback or try/catch for scenarios that cannot happen.
  • No trivial comments that restate the code.

Files in this skill

Load the file that matches your task. Do not load them all at once.

  • getting-started.md five-minute install and first screen. Read this first if Forge is not already wired up.
  • components.md every export grouped by category, with props table. Read this when picking a component or looking up props.
  • tokens.md CSS variables, spacing scale, radius, shadows, z-index, motion tokens, breakpoints, color constants. Read this when styling edges Forge does not cover with a prop.
  • theming.md ForgeProvider props, ForgeTheme fields, default darkTheme and lightTheme, createTheme, useForge. Read this when configuring the provider or building a custom theme.
  • motion.md the Motion library (Motion, AnimatePresence, Stagger, gesture, scroll, text/visual effects, canvas backgrounds, motion hooks). Read this when adding animations.
  • patterns.md full compositions for dashboards, forms, responsive sidebars, list-with-actions, plus quality gates per pattern and block creation rules. Read this when building a page, not a single component.
  • a11y.md accessibility playbook: what Forge adds automatically per component and what the dev must still add. Read this when working on any interactive surface.
  • anti-patterns.md wrong/right pairs for the most common mistakes. Skim this when you are about to write a styled native element, a flex div, a @media query, or a hardcoded color.
  • design.md design rules and the reasoning behind them. Read this when choosing between multiple valid approaches or when the user asks "why".
  • migration.md upgrade notes from previous versions. Read this when bumping wss3-forge.
  • faq.md common gotchas and their fixes (unstyled components, 404s, wrong sizes). Read this when something does not behave as expected.

Typical flow

  1. Read the user's request.
  2. Open components.md and find the primitive that matches. If the request is a composition, open patterns.md first.
  3. If the primitive needs a color, radius, or spacing you cannot express with a prop, open tokens.md.
  4. If the request involves animation, open motion.md.
  5. If the request touches theming or provider configuration, open theming.md.
  6. If the request is interactive (form, menu, dialog, table), skim a11y.md.
  7. Write the code. Run through the Review priority and Pre-delivery checklist. Run tsc --noEmit before claiming done.

Blocks

Block templates live under src/pages/blocks/blocks/{category}/{slug}/. Categories: marketing | dashboard | auth | commerce | settings | content | utility. Block creation rules are documented in patterns.md.

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 Forge AI skill do?

Forge design system for React. Use when writing any React UI in this project, when picking which component to use, when theming, when creating block templates, or when reviewing UI code for accessibility and anti-patterns. Covers components, tokens, theming, motion, patterns, a11y, and anti-patterns.

Why use Forge on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Webba-Creative-Technologies/forge/tree/main/skills/forge. 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 Forge?

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 Forge?

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

Is the Forge AI skill free?

Yes. It is published on GitHub by Webba-Creative-Technologies 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 👇