Ape Design logo

Ape Design

Community
arpitbbhayani
ape-design

Enforces the Array UI (array-ui) editorial systems design system for any frontend, UI, website, blog, personal site, SaaS dashboard, or documentation development task. Trigger on "ape design", "Array UI", or whenever building, styling, or scaffolding UI to guarantee high information density, mathematical precision, editorial typography, and 0% custom CSS without prompt loops.

Overview

Publisherarpitbbhayani
Repositoryape-skills
Skill nameape-design
Stars
51
Forks
5
Bundled files
3
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.

  • 3 bundled files

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

  • Open source

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

Installation

Install the Ape Design 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/arpitbbhayani/ape-skills/tree/master/ape-design
  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/arpitbbhayani/ape-skills.git /tmp/ape-skills
mkdir -p .claude/skills
cp -r /tmp/ape-skills/ape-design .claude/skills/ape-design
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ape Design 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 Ape Design 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 Ape Design 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.

ape-design: Editorial Systems Design Skill (Array UI)

This skill enforces strict adherence to Array UI (array-ui)—the editorial, systems-engineering design system inspired by arpitbhayani.me and distributed at ui.arpitbhayani.me.

When building or modifying any website (personal portfolio, engineering blog, SaaS dashboard, technical documentation, or systems explainer), the agent MUST follow this workflow and rules without requiring prompt corrections.


1. The 6 Non-Negotiable Prime Invariants

Invariant 1: Mandatory Font Enforcement (DO THIS FIRST)

[!IMPORTANT] Before creating or editing any components, the agent MUST verify that Google Fonts are loaded and applied to body. If fonts are missing, the UI will fall back to system sans-serif (Arial/Inter), completely breaking the editorial authority.

  1. Google Fonts URL: https://fonts.googleapis.com/css2?family=Assistant:wght@400;500;600;700;800&family=IBM+Plex+Mono:ital,wght@0,400;0,500;0,600;0,700;1,400&family=Lora:ital,wght@0,400..700;1,400..700&display=swap
  2. In globals.css (or main CSS):
    css
    @import url("https://fonts.googleapis.com/css2?family=Assistant:wght@400;500;600;700;800&family=IBM+Plex+Mono:ital,wght@0,400;0,500;0,600;0,700;1,400&family=Lora:ital,wght@0,400..700;1,400..700&display=swap");
    
    body {
      font-family: var(--font-sans, "Assistant", -apple-system, sans-serif);
      -webkit-font-smoothing: antialiased;
    }
  3. Or in root layout <head>: Include <ThemeScript fonts /> from @aui/theme-script.

Invariant 2: 0% Custom CSS Rule

[!CAUTION] Custom CSS proportion MUST REMAIN 0% AT ALL TIMES.

  • NEVER create custom .css files.
  • NEVER write <style> blocks or style tags in .astro, .tsx, .jsx, or .html files.
  • NEVER inject arbitrary ad-hoc inline styles for margins, paddings, colors, or shadows.
  • All layout, typography, blank spacing, telemetry, and interactive controls MUST be composed exclusively using native Array UI components (PageShell, Navbar, Footer, Card, Button, Table, Badge, Terminal, PingStatus, DiffBlock, Gutter, Layout, etc.) and system utility classes (.aui-grid, .aui-stack, .aui-row, .aui-container).

Invariant 3: Frame Every Page in <PageShell>

Never hand-roll outer page containers or guess margin widths. Wrap every full page in:

tsx
<PageShell navbar={<Navbar ... />} footer={<Footer ... />}>
  {/* Page content */}
</PageShell>

PageShell (@aui/page-shell) automatically provides a centered max-w-6xl column, proper horizontal padding, full viewport height (min-h-screen flex flex-col), and clean baseline typography.

Invariant 4: Editorial Footer Discipline

The footer must match arpitbhayani.me exactly:

  • Canvas: Flat paper (#faf9f5) or obsidian carbon (#111113) canvas (bg-background). Never use floating tinted card backgrounds (bg-card/60) for the footer.
  • Hairlines: Exactly 1px border-t border-border on top, and border-t border-border/70 separating disclaimer and bottom bar.
  • Link Layout: Links are inline items (inline-flex items-center gap-1.5) with hover crimson (text-primary) and subtle translate-x-0.5. Hints are rendered inline in subtle monospace (text-[11px] font-mono text-muted-foreground/60 like Projects (Open-source)). NEVER use justify-between with boxed badge pills stretching across the column.
  • Bottom Bar: Copyright on the left, tactile <SocialPill>s with tooltips on the right.

Invariant 5: Wrap Every Action & Status in <Tooltip>

  • Every action button, icon trigger, theme toggle, copy button, and <PingStatus> MUST be wrapped in a <Tooltip content="...">.
  • Communicate consequence, keyboard shortcuts, or live telemetry facts (e.g. <Tooltip content="Toggle dark / light mode (⌘D)">).

Invariant 6: Anti-Card-Soup & Visual Rhythm

  • NEVER stack identical layout blocks consecutively (e.g. never place two 3-card grids in a row, or three identical text callouts sequentially).
  • Alternate visual weight: Pair dense, machined telemetry (Terminal, DiffBlock, ParamTable, PropertyGrid) with airy, editorial moments (Maxim, pull quotes, <Gutter size="xl" />, TakeawaysBox).
  • Use asymmetric composition: Prefer 2-column asymmetric splits (SplitPane, 60/40 hero-to-terminal layouts) over monotonous symmetric grids.

2. Palette & Visual Standards

  1. Editorial Palette:

    • Light mode: Paper canvas (#faf9f5), warm hairline borders (#dfded9), dark carbon text (#2b2a30).
    • Dark mode: Obsidian carbon (#111113), crisp border (#27272c), soft readable white (#d6d6dc).
    • Signature Crimson (#e5000f in light, #ff3344 in dark): Reserved strictly for primary action buttons, active navigation indicators, and key focus rings.
    • Amber (#cc9900): Used exclusively for engineering maxim quotes, philosophy callouts, and warning banners.
    • Emerald (#10b981): Used for operational health, ping dots, uptime, and git diff additions.
    • Rose (#f43f5e): Used for errors, outages, and git diff deletions.
  2. Typography:

    • Headings & Body: Assistant (dense, disciplined, 15px base size, tracking -0.015em on headings).
    • Pull Quotes / Maxims: Lora (italic serif).
    • Code & Telemetry: IBM Plex Mono / system monospace.
  3. Restraint & Elevation:

    • 1px hairline borders everywhere.
    • Flat, tinted card surfaces (bg-muted/30 or var(--aui-bg-secondary)).
    • Zero heavy drop shadows.

3. Installation & Autonomous Execution

The agent MUST NOT ask the user which components to install. Autonomously inspect what is needed and run:

Mode A: shadcn CLI (Recommended for Next.js, Vite, Tailwind)

  1. Add registry to components.json:
    json
    {
      "$schema": "https://ui.shadcn.com/schema.json",
      "style": "default",
      "registries": {
        "@aui": "https://ui.arpitbhayani.me/r/{name}.json"
      }
    }
  2. Install theme & base layer:
    bash
    npx shadcn@latest add @aui/aui -y
  3. Install required components autonomously with -y:
    bash
    npx shadcn@latest add @aui/page-shell @aui/navbar @aui/footer @aui/button @aui/badge @aui/card @aui/terminal @aui/diff-block @aui/ping-status @aui/tooltip @aui/gutter @aui/layout -y

Mode B: Direct Package Dependency (Next.js, Astro)

bash
npm install array-ui

Import styles in root layout:

tsx
import "array-ui/styles.css";

4. Production Layout Recipes

Recipe 1: PageShell & Global Navigation Scaffold

tsx
import { PageShell } from "@/components/ui/page-shell";
import { Navbar } from "@/components/ui/navbar";
import { Footer } from "@/components/ui/footer";
import { Button } from "@/components/ui/button";
import { Tooltip } from "@/components/ui/tooltip";
import { PingStatus } from "@/components/ui/ping-status";
import { ThemeToggle } from "@/components/ui/theme-toggle";
import { SocialPill } from "@/components/ui/social-pill";

export function SiteShell({ children }: { children: React.ReactNode }) {
  return (
    <PageShell
      navbar={
        <Navbar
          brandName="VectorMesh"
          brandHref="/"
          links={[
            { label: "Architecture", href: "/architecture" },
            { label: "Telemetry", href: "/telemetry" },
            { label: "Writings", href: "/writings" },
            { label: "Docs", href: "/docs" },
          ]}
          statusIndicator={
            <Tooltip content="All 16 quorum nodes operational · p99 < 12ms">
              <PingStatus status="operational" label="99.99% Live" size="sm" />
            </Tooltip>
          }
          actions={
            <div className="flex items-center gap-2">
              <Tooltip content="Toggle theme (⌘D)">
                <ThemeToggle variant="ghost" size="sm" />
              </Tooltip>
              <Tooltip content="Launch interactive browser cluster">
                <Button variant="primary" size="sm">
                  Deploy
                </Button>
              </Tooltip>
            </div>
          }
        />
      }
      footer={
        <Footer
          columns={[
            {
              heading: "Engine",
              links: [
                { label: "Consensus", href: "/consensus", hint: "Raft" },
                { label: "Storage Engine", href: "/storage", hint: "LSM" },
                { label: "WAL", href: "/wal" },
              ],
            },
            {
              heading: "Resources",
              links: [
                { label: "Documentation", href: "/docs" },
                { label: "API Reference", href: "/api", hint: "v2.4" },
                { label: "Benchmarks", href: "/benchmarks" },
              ],
            },
            {
              heading: "Open Source",
              links: [
                { label: "GitHub", href: "https://github.com", external: true },
                { label: "RFCs", href: "/rfcs" },
                { label: "Changelog", href: "/changelog" },
              ],
            },
          ]}
          disclaimer="VectorMesh is an open-source distributed log replication engine built for zero-copy streaming workloads. MIT Licensed."
          copyright={`© ${new Date().getFullYear()} VectorMesh. Built for curious engineers.`}
          socialPills={
            <>
              <Tooltip content="Watch deep-dives on YouTube">
                <SocialPill platform="youtube" href="https://youtube.com" count="210k" />
              </Tooltip>
              <Tooltip content="Follow engineering on X">
                <SocialPill platform="twitter" href="https://twitter.com" count="120k" />
              </Tooltip>
              <Tooltip content="Star on GitHub">
                <SocialPill platform="github" href="https://github.com" count="7k" />
              </Tooltip>
            </>
          }
        />
      }
    >
      {children}
    </PageShell>
  );
}

Recipe 2: High-Converting Technical Landing Page

tsx
import { Gutter } from "@/components/ui/gutter";
import { Container, Stack, Row, Grid } from "@/components/ui/layout";
import { Button } from "@/components/ui/button";
import { Badge } from "@/components/ui/badge";
import { Card, CardHeader, CardTitle, CardContent } from "@/components/ui/card";
import { Terminal } from "@/components/ui/terminal";
import { Maxim } from "@/components/ui/maxim";
import { Tooltip } from "@/components/ui/tooltip";
import { StatCard } from "@/components/ui/stat-card";

export default function LandingPage() {
  return (
    <Stack gap="xl">
      {/* 1. Top Announcement Nudge */}
      <Row justify="start">
        <Tooltip content="Read the v2.4 zero-copy streaming benchmark report">
          <Badge variant="amber" interactive href="/changelog">
            v2.4 Released: Zero-copy disk streaming engine is live →
          </Badge>
        </Tooltip>
      </Row>

      {/* 2. Hero Proposition */}
      <Stack gap="md" className="max-w-3xl">
        <h1 className="text-4xl sm:text-5xl lg:text-6xl font-bold font-heading tracking-tight leading-[1.08]">
          Deterministic consensus at <span className="text-primary">hardware speed</span>.
        </h1>
        <p className="text-lg text-muted-foreground leading-relaxed font-sans">
          A high-density distributed write-ahead log with strict Raft quorum, zero GC pauses, and sub-millisecond p99 replication latency.
        </p>

        {/* Action CTAs wrapped in Tooltips */}
        <Row gap="sm" align="center" wrap>
          <Tooltip content="Launches live interactive telemetry cluster (zero signup required)">
            <Button variant="primary" size="lg">
              Launch Sandbox Quorum
            </Button>
          </Tooltip>
          <Tooltip content="Read architectural whitepaper and formal TLA+ proofs">
            <Button variant="secondary" size="lg">
              Read Architecture RFC →
            </Button>
          </Tooltip>
        </Row>

        {/* Friction-Reduction Reassurance Nudge Row */}
        <Row gap="xs" align="center" wrap>
          <Badge variant="secondary">Zero external dependencies</Badge>
          <Badge variant="green">Apache 2.0</Badge>
          <Badge variant="blue">Single static binary</Badge>
          <Badge variant="secondary">Sub-10ms failover</Badge>
        </Row>
      </Stack>

      <Gutter size="md" />

      {/* 3. The Live Mechanical Proof (Terminal Breakout) */}
      <Terminal
        title="~/vectormesh (cluster: eu-west-1)"
        lines={[
          "vmesh cluster bootstrap --peers 3 --wal-dir /data/wal",
          { text: "✔ Node 1 [Leader]: Quorum established (term=42, index=108420)", kind: "ok" },
          { text: "✔ Node 2 [Follower]: Log synchronized in 0.84ms", kind: "ok" },
          { text: "✔ Node 3 [Follower]: Heartbeat acknowledged, fsync verified", kind: "ok" },
          { text: "Cluster status: HEALTHY · Throughput: 420,000 ops/sec", kind: "out" },
        ]}
      />

      <Gutter size="lg" />

      {/* 4. Concrete Metric Pillars */}
      <Grid cols={3} gap="md">
        <StatCard value="420k" label="ops / second" description="sustained write throughput" accent="emerald" />
        <StatCard value="0.84ms" label="p99 replication" description="cross-region Raft quorum" accent="blue" />
        <StatCard value="Zero" label="memory allocations" description="ring-buffer page cache" accent="rose" />
      </Grid>

      <Gutter size="lg" />

      {/* 5. Asymmetric Deep-Dive Cards */}
      <Grid cols={2} gap="lg">
        <Card hoverable>
          <CardHeader>
            <Badge variant="violet" className="w-fit mb-2">Storage Engine</Badge>
            <CardTitle>Memory-Mapped Append Log</CardTitle>
          </CardHeader>
          <CardContent className="text-muted-foreground text-sm leading-relaxed">
            Direct I/O bypasses OS page cache jitter. Writes are sequentially appended to 64MB pre-allocated segments with hardware-enforced crc32 checksum verification.
          </CardContent>
        </Card>

        <Card hoverable>
          <CardHeader>
            <Badge variant="cyan" className="w-fit mb-2">Consensus</Badge>
            <CardTitle>Pipelined Raft Replication</CardTitle>
          </CardHeader>
          <CardContent className="text-muted-foreground text-sm leading-relaxed">
            Entries are optimistically transmitted to quorum followers in lock-free ring buffers while the leader executes fsync concurrently, eliminating network round-trip bottlenecks.
          </CardContent>
        </Card>
      </Grid>

      <Gutter size="lg" />

      {/* 6. Engineering Maxim Quote */}
      <Maxim author="Arpit Bhayani" source="Asli Engineering">
        Simplicity is a prerequisite for reliability. Complex systems fail in complex, unexpected ways.
      </Maxim>
    </Stack>
  );
}

5. Self-Audit Checklist (Verify Before Concluding)

Before declaring any UI task done, the agent MUST verify every item:

  • Fonts Enforced: Google Fonts (Assistant, IBM Plex Mono, Lora) loaded in globals.css or <head>, and body { font-family: var(--font-sans) } active.
  • 0% Custom CSS: Zero .css files created, zero <style> blocks written, zero ad-hoc margin/padding styles.
  • PageShell Enforced: Screen is framed in <PageShell> with centered max-w-6xl bounds.
  • Footer Editorial: Footer is flat on bg-background with hairline border, inline links with (hint), and social pills.
  • Tooltips Wrapped: Every action button, theme toggle, copy button, and <PingStatus> has a <Tooltip>.
  • Visual Rhythm: No consecutive stacks of identical cards; telemetry breakouts and maxims alternate with prose.

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 Ape Design AI skill do?

Enforces the Array UI (array-ui) editorial systems design system for any frontend, UI, website, blog, personal site, SaaS dashboard, or documentation development task. Trigger on "ape design", "Array UI", or whenever building, styling, or scaffolding UI to guarantee high information density, mathematical precision, editorial typography, and 0% custom CSS without prompt loops.

Why use Ape Design on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/arpitbbhayani/ape-skills/tree/master/ape-design. 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 Ape Design?

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 Ape Design?

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

Is the Ape Design AI skill free?

Yes. It is published on GitHub by arpitbbhayani 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 👇