Fitts Law Touch Targets logo

Fitts Law Touch Targets

Community
Deibler
fitts-law-touch-targets

Use this skill when designing for touch — phones, tablets, kiosks, in-car displays, smart-TV remotes, anything where the input is a finger or thumb rather than a precise pointer. Trigger when sizing buttons for mobile, debugging "users keep mistapping," designing a mobile bottom-nav, picking icon sizes for a phone toolbar, or reviewing whether a desktop UI is touch-friendly. Sub-aspect of `fitts-law`; read that first if you haven't already.

Overview

PublisherDeibler
Repositoryuniversal-design-principles
Skill namefitts-law-touch-targets
Stars
9
Forks
3
Bundled files
1
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 Deibler on GitHub. Read the source before you install it.

Installation

Install the Fitts Law Touch Targets 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/Deibler/universal-design-principles/tree/main/plugins/interaction-and-control-principles/skills/fitts-law-touch-targets
  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/Deibler/universal-design-principles.git /tmp/universal-design-principles
mkdir -p .claude/skills
cp -r /tmp/universal-design-principles/plugins/interaction-and-control-principles/skills/fitts-law-touch-targets .claude/skills/fitts-law-touch-targets
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Fitts Law Touch Targets 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 Fitts Law Touch Targets 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 Fitts Law Touch Targets 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.

Fitts's Law for touch targets

The finger is a much wider, much less precise pointer than a mouse cursor. Fitts's Law still applies, but the constants change dramatically — and a few touch-specific phenomena (occlusion, thumb reach, edge gestures) layer on top. This skill covers the practical rules.

Minimum sizes

Every major touch platform publishes a recommended minimum. They converge:

Platform / specMinimum targetNotes
Apple HIG (iOS)44×44 ptThe most-cited number; matches average finger contact.
Material Design (Android)48×48 dpRoughly equivalent to Apple's 44pt at common densities.
Microsoft (Windows touch)40×40 effective pxSlightly smaller; Microsoft's research basis.
WCAG 2.5.5 (Level AAA)44×44 CSS pxStrictest accessibility minimum.
WCAG 2.5.8 (Level AA, 2.2)24×24 CSS px (with adequate spacing)Newer, more lenient minimum.

The practical rule: 44×44 CSS pixels for any touchable element. This is the safe number across platforms and accessibility levels. Going smaller is occasionally defensible in specific dense expert contexts; never as a default.

The visible icon inside the button can be smaller (16–24px is common). The hit area — the actual touchable region — is what 44×44 measures.

css
/* Right: a small icon with a generous tap area */
.icon-button {
  display: inline-grid;
  place-items: center;
  width: 44px;
  height: 44px;
  padding: 10px;        /* keeps icon centered while tap area stays at 44 */
  border: 0;
  background: transparent;
  border-radius: 22px;
}
.icon-button svg { width: 20px; height: 20px; }

Spacing between targets

Fitts's Law is about acquiring one target. When targets are clustered, spacing between them prevents accidental neighbor-taps. The convention:

  • At least 8px of clear space between adjacent touch targets.
  • 44×44 hit area applies to each target, even if visually they overlap (e.g., padded buttons share visual edges).

For dense clusters (chip rows, segmented controls), increase visual differentiation: bordered, with internal padding, with adequate gap.

The thumb-reach map

When a phone is held one-handed, the thumb can reach some screen regions easily and others only with strain. Research (Hoober and others) maps this:

+--------------------+
|       hard         |  <- top of screen, far from thumb base
|   hard | medium    |
| medium | easy      |
| medium | easy      |
| medium | easy      |  <- bottom-right (right hand) is easiest
+--------------------+

For right-handed users, the bottom-right of the screen is easiest; the top-left is hardest. Left-handed users mirror.

Practical implications:

  • Primary actions go in the lower portion of the screen. Floating action buttons (FAB) at bottom-right are a strong Fitts-and-thumb-reach win.
  • Bottom navigation beats top navigation for high-frequency tap-to-navigate patterns.
  • Top-bar actions should be infrequent or hand-agnostic (search, settings, profile menu — not "save" or "publish").
  • Reachability features (iOS double-tap-home; Android one-handed mode) exist because top corners are genuinely hard.
html
<!-- Bottom nav: thumb-friendly, primary destinations -->
<nav style="position: fixed; bottom: 0; left: 0; right: 0;">
  <!-- 3-5 destinations, evenly spaced, 56px+ tall -->
</nav>

<!-- FAB: bottom-right for right-handed primary action -->
<button style="position: fixed; bottom: 24px; right: 24px; width: 56px; height: 56px; border-radius: 28px;">
  <PlusIcon />
</button>

Occlusion (the finger covers the target)

Once the user's finger touches the screen, it covers the target. A few consequences:

  • Confirm visually after touch, not just on touch-up. The user wants to know the press registered before they lift.
  • Don't rely on the user reading text under their finger. Tooltips, micro-confirmations, "what does this do?" hints under the touch point are invisible.
  • Touch targets should provide visual state changes that radiate beyond the finger — color flash on the whole row, a ripple effect, an outline that extends past the contact area.

Material Design's ripple effect is a classic occlusion-mitigation: the ripple expands outward from the touch point, providing visible feedback that survives the finger's presence.

css
@keyframes ripple {
  to { transform: scale(4); opacity: 0; }
}
.button-pressed::after {
  content: '';
  position: absolute;
  border-radius: 50%;
  background: currentColor;
  opacity: 0.2;
  animation: ripple 600ms;
}

Edge gestures and the safe area

Modern phones use the screen edges for system gestures (back-swipe, app-switcher, control-center). A button placed at the absolute edge competes with these gestures.

Practical rule: keep an 8–16px inset from the edges for primary interactive content. Use the system's safe-area insets for top notch and bottom home-indicator clearance:

css
.bottom-bar {
  padding-bottom: env(safe-area-inset-bottom);
}
.top-bar {
  padding-top: env(safe-area-inset-top);
}

The screen-edge Fitts-Law advantage still works for layout (a tap target abutting the edge is large in one dimension), but interaction must respect system gestures.

Hover states (and why they don't translate)

Touch devices have no hover. UI patterns that rely on hover-revealed actions (a row's "Edit" / "Delete" buttons that appear on cursor over) silently fail on touch. Mitigations:

  • Always-visible row actions (perhaps dimmed; see below). Touch users see and can tap them.
  • Press-and-hold to reveal a context menu (analogous to right-click).
  • Swipe gestures for revealed actions (iOS-style swipe-to-delete on a list row).

Don't ship "hover to discover the action" patterns to mobile.

Worked examples

Example 1: a row in a list with an action button

Anti-pattern (mouse-only):

html
<li class="list-item">
  <span>Project Acme</span>
  <button class="hover-only" style="opacity: 0;">Open</button>
</li>
<style>
  .list-item:hover .hover-only { opacity: 1; }
</style>

On touch, the button is invisible — and even if you tap it, you can't see it.

Right (touch-friendly):

html
<li class="list-item" style="display: flex; align-items: center; padding: 12px 16px; min-height: 56px;">
  <span style="flex: 1;">Project Acme</span>
  <button aria-label="Open Acme" style="width: 44px; height: 44px;">
    <ChevronRightIcon />
  </button>
</li>

The action is always visible; the row itself meets the 44px height minimum (56px is even better).

Example 2: a numeric stepper

html
<div class="stepper" style="display: inline-flex; align-items: center; gap: 12px;">
  <button aria-label="Decrease" style="width: 44px; height: 44px;">−</button>
  <input type="number" value="1" style="width: 60px; height: 44px; text-align: center;" />
  <button aria-label="Increase" style="width: 44px; height: 44px;">+</button>
</div>

Both buttons are 44×44 with 12px gap to the input — generous spacing prevents misclicks.

Example 3: a floating action button (FAB)

html
<button class="fab" aria-label="New post" style="
  position: fixed;
  bottom: calc(24px + env(safe-area-inset-bottom));
  right: 24px;
  width: 56px; height: 56px;
  border-radius: 28px;
  background: var(--brand);
  color: white;
  display: grid; place-items: center;
  box-shadow: 0 4px 12px rgba(0,0,0,0.15);
">
  <PlusIcon />
</button>

Bottom-right for thumb reach. 56×56 hit area. Safe-area-inset for home-indicator clearance. Subtle shadow because FABs do float above content.

Example 4: a tab bar

html
<nav class="tab-bar" style="
  position: fixed; bottom: 0; left: 0; right: 0;
  display: grid; grid-template-columns: repeat(5, 1fr);
  height: calc(56px + env(safe-area-inset-bottom));
  padding-bottom: env(safe-area-inset-bottom);
  background: white;
  border-top: 1px solid hsl(0 0% 90%);
">
  <button class="tab" aria-current="page">
    <HomeIcon style="width: 24px; height: 24px;" />
    <span style="font-size: 10px;">Home</span>
  </button>
  <!-- four more tabs -->
</nav>

<style>
  .tab {
    display: grid; place-items: center; gap: 2px;
    border: 0; background: transparent;
    color: hsl(0 0% 50%);
  }
  .tab[aria-current="page"] { color: var(--brand); }
</style>

5 tabs across the bottom. Each tab gets ~76px width × 56px height — generously above 44×44.

Anti-patterns

  • Tiny X close on a popover. A 16×16 X in the corner. On touch, nearly impossible to hit reliably.
  • Stacked menu items at 32px row height. The user means to tap "Settings" and gets "Sign out." Make rows ≥ 44px (preferably 48–56px) on touch.
  • Edge-flush primary actions. A "Send" button at the absolute right edge that triggers the system back-gesture instead. Inset 8–16px.
  • Hover-only actions. Row actions, contextual menus, tooltips that only appear on cursor-over. Either always-visible or accessible via tap-to-reveal.
  • Inert spacing. A 44×44 visual button with only 16×16 of actual touchable area (because some CSS issue). Audit with browser DevTools' touch emulator.

Heuristics

  1. The thumb test. Hold your phone one-handed, naturally. Try to tap every primary action. Anything that requires a hand-shift is wrong-position.
  2. The fingertip overlay. In design comps, overlay a 44×44 circle on every interactive element. Anything where the circle exceeds the visual element's bounds is fine; anything smaller than the circle is too small.
  3. Real-device testing on the smallest target screen. A design that works on a Pixel 7 may fail on a small Android device or iPhone SE. Test on actual hardware in actual hands.
  4. The bystander test. Watch someone else use your app. Where do they hesitate or correct themselves? Each correction is a Fitts-or-occlusion failure.

Related sub-skills

  • fitts-law (parent).
  • fitts-law-screen-edges — companion for desktop; some overlap on mobile (safe areas).
  • affordance (parent plugin) — touch targets need visual affordance to be discoverable.
  • accessibility-operable (process plugin) — touch target sizes are an accessibility minimum; don't go below.

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 Fitts Law Touch Targets AI skill do?

Use this skill when designing for touch — phones, tablets, kiosks, in-car displays, smart-TV remotes, anything where the input is a finger or thumb rather than a precise pointer. Trigger when sizing buttons for mobile, debugging "users keep mistapping," designing a mobile bottom-nav, picking icon sizes for a phone toolbar, or reviewing whether a desktop UI is touch-friendly. Sub-aspect of `fitts-law`; read that first if you haven't already.

Why use Fitts Law Touch Targets on TypingMind?

Because you install it once and use it with any model. Fitts Law Touch Targets 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 Fitts Law Touch Targets in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Deibler/universal-design-principles/tree/main/plugins/interaction-and-control-principles/skills/fitts-law-touch-targets. 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 Fitts Law Touch Targets?

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 Fitts Law Touch Targets?

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

Is the Fitts Law Touch Targets AI skill free?

It is published on GitHub by Deibler. Check the repository for licensing terms. 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 👇