Best Practices logo

Best Practices

CommunityPopular
addyosmani
best-practices

Apply modern web development best practices for security, compatibility, and code quality. Use when asked to "apply best practices", "security audit", "modernize code", "code quality review", or "check for vulnerabilities".

Overview

Publisheraddyosmani
Repositoryweb-quality-skills
Skill namebest-practices
Stars
2.8K
Forks
246
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 addyosmani on GitHub. Read the source before you install it.

Installation

Install the Best Practices 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.
  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/addyosmani/web-quality-skills.git /tmp/web-quality-skills
mkdir -p .claude/skills
cp -r /tmp/web-quality-skills/skills/best-practices .claude/skills/best-practices
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Best Practices 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 Best Practices 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 Best Practices 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.

Best practices

Modern web development standards based on Lighthouse best practices audits. Covers security, browser compatibility, and code quality patterns.

Evidence-led audit workflow

When a rendered page is available:

  1. Run a live Lighthouse Best Practices audit when that capability is available; with Chrome DevTools MCP, use lighthouse_audit. Use navigation mode for a normal page load or snapshot mode when the current state must be preserved.
  2. Inspect the listed console and network failures and fetch individual details only when they support a finding.
  3. Supplement runtime evidence with dependency, header, configuration, and source inspection; Lighthouse is not a complete security assessment.
  4. Fix the implicated code, re-run the same audit, and keep security findings separate from style preferences.

If live tools are unavailable, use the Lighthouse CLI plus focused dependency and header checks. Never report a high Lighthouse score as proof that the application is secure.

Security

Read the security reference when security is in scope or a live audit surfaces a related failure. It covers HTTPS/HSTS, CSP and Trusted Types, Subresource Integrity, headers, dependencies, sanitization, and cookies.

At minimum:

  • Use HTTPS without mixed content. Add HSTS only after confirming every relevant subdomain supports HTTPS.
  • Treat a strict CSP as defense in depth. Prefer nonces or hashes and test with report-only before enforcement.
  • Sanitize untrusted HTML and protect DOM XSS sinks. Prefer text APIs when markup is not required.
  • Pin and review third-party code. Use SRI where the delivery model supports it and keep dependencies patched.
  • Verify response headers at runtime. Source configuration alone does not prove what the deployed page sends.

Browser compatibility

Doctype declaration

html
<!-- ❌ Missing or invalid doctype -->
<HTML>
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN">

<!-- ✅ HTML5 doctype -->
<!DOCTYPE html>
<html lang="en">

Character encoding

html
<!-- ❌ Missing or late charset -->
<html>
<head>
  <title>Page</title>
  <meta charset="UTF-8">
</head>

<!-- ✅ Charset as first element in head -->
<html>
<head>
  <meta charset="UTF-8">
  <title>Page</title>
</head>

Viewport meta tag

html
<!-- ❌ Missing viewport -->
<head>
  <title>Page</title>
</head>

<!-- ✅ Responsive viewport -->
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1">
  <title>Page</title>
</head>

Feature detection

javascript
// ❌ Browser detection (brittle)
if (navigator.userAgent.includes('Chrome')) {
  // Chrome-specific code
}

// ✅ Feature detection
if ('IntersectionObserver' in window) {
  // Use IntersectionObserver
} else {
  // Fallback
}

// ✅ Using @supports in CSS
@supports (display: grid) {
  .container {
    display: grid;
  }
}

@supports not (display: grid) {
  .container {
    display: flex;
  }
}

Polyfills (when needed)

Prefer bundling polyfills at build time (Babel/SWC + core-js, or @vitejs/plugin-legacy) targeted by your supported-browsers list. This eliminates the runtime check entirely and avoids shipping polyfill bytes to modern browsers.

If you must load a polyfill at runtime, append a script element — never use document.write (it blocks the parser and is broken in async/deferred contexts):

html
<script>
  if (!('fetch' in window)) {
    const s = document.createElement('script');
    s.src = '/polyfills/fetch.js';
    s.defer = true;
    document.head.appendChild(s);
  }
</script>

Never load polyfills from a third-party CDN you don't control. The polyfill.io service was compromised in mid-2024 in a supply-chain attack and used to serve malware to ~100k sites. Self-host, or use a vetted mirror (e.g. Cloudflare's cdnjs polyfill build) — and pin the version with Subresource Integrity.


Deprecated APIs

Avoid these

javascript
// ❌ document.write (blocks parsing)
document.write('<script src="..."></script>');

// ✅ Dynamic script loading
const script = document.createElement('script');
script.src = '...';
document.head.appendChild(script);

// ❌ Synchronous XHR (blocks main thread)
const xhr = new XMLHttpRequest();
xhr.open('GET', url, false); // false = synchronous

// ✅ Async fetch
const response = await fetch(url);

// ❌ Application Cache (deprecated)
<html manifest="cache.manifest">

// ✅ Service Workers
if ('serviceWorker' in navigator) {
  navigator.serviceWorker.register('/sw.js');
}

Event listener passive

javascript
// ❌ Non-passive touch/wheel (may block scrolling)
element.addEventListener('touchstart', handler);
element.addEventListener('wheel', handler);

// ✅ Passive listeners (allows smooth scrolling)
element.addEventListener('touchstart', handler, { passive: true });
element.addEventListener('wheel', handler, { passive: true });

// ✅ If you need preventDefault, be explicit
element.addEventListener('touchstart', handler, { passive: false });

Console & errors

No console errors

javascript
// ❌ Errors in production
console.log('Debug info'); // Remove in production
throw new Error('Unhandled'); // Catch all errors

// ✅ Proper error handling
try {
  riskyOperation();
} catch (error) {
  // Log to error tracking service
  errorTracker.captureException(error);
  // Show user-friendly message
  showErrorMessage('Something went wrong. Please try again.');
}

Error boundaries (React)

jsx
class ErrorBoundary extends React.Component {
  state = { hasError: false };
  
  static getDerivedStateFromError(error) {
    return { hasError: true };
  }
  
  componentDidCatch(error, info) {
    errorTracker.captureException(error, { extra: info });
  }
  
  render() {
    if (this.state.hasError) {
      return <FallbackUI />;
    }
    return this.props.children;
  }
}

// Usage
<ErrorBoundary>
  <App />
</ErrorBoundary>

Global error handler

javascript
// Catch unhandled errors
window.addEventListener('error', (event) => {
  errorTracker.captureException(event.error);
});

// Catch unhandled promise rejections
window.addEventListener('unhandledrejection', (event) => {
  errorTracker.captureException(event.reason);
});

Source maps

Production configuration

javascript
// ❌ Source maps exposed in production
// webpack.config.js
module.exports = {
  devtool: 'source-map', // Exposes source code
};

// ✅ Hidden source maps (uploaded to error tracker)
module.exports = {
  devtool: 'hidden-source-map',
};

// ✅ Or no source maps in production
module.exports = {
  devtool: process.env.NODE_ENV === 'production' ? false : 'source-map',
};

Strip sourcesContent from production maps when uploading to your error tracker. By default, bundlers embed the full original source inside the .map file — anyone who obtains the map (including via a misconfigured upload step) gets your unminified code. Configure your bundler to omit sourcesContent, or use a Sentry/Bugsnag CLI flag that does so when uploading.

For Vite, prefer sourcemap: 'hidden' over 'true' so the //# sourceMappingURL= comment isn't emitted into the bundle.


Performance best practices

Avoid blocking patterns

javascript
// ❌ Blocking script
<script src="heavy-library.js"></script>

// ✅ Deferred script
<script defer src="heavy-library.js"></script>

// ❌ Blocking CSS import
@import url('other-styles.css');

// ✅ Link tags (parallel loading)
<link rel="stylesheet" href="styles.css">
<link rel="stylesheet" href="other-styles.css">

Efficient event handlers

javascript
// ❌ Handler on every element
items.forEach(item => {
  item.addEventListener('click', handleClick);
});

// ✅ Event delegation
container.addEventListener('click', (e) => {
  if (e.target.matches('.item')) {
    handleClick(e);
  }
});

Memory management

javascript
// ❌ Memory leak (never removed)
const handler = () => { /* ... */ };
window.addEventListener('resize', handler);

// ✅ Cleanup when done
const handler = () => { /* ... */ };
window.addEventListener('resize', handler);

// Later, when component unmounts:
window.removeEventListener('resize', handler);

// ✅ Using AbortController
const controller = new AbortController();
window.addEventListener('resize', handler, { signal: controller.signal });

// Cleanup:
controller.abort();

Code quality

Valid HTML

html
<!-- ❌ Invalid HTML -->
<div id="header">
<div id="header"> <!-- Duplicate ID -->

<ul>
  <div>Item</div> <!-- Invalid child -->
</ul>

<a href="/"><button>Click</button></a> <!-- Invalid nesting -->

<!-- ✅ Valid HTML -->
<header id="site-header">
</header>

<ul>
  <li>Item</li>
</ul>

<a href="/" class="button">Click</a>

Semantic HTML

html
<!-- ❌ Non-semantic -->
<div class="header">
  <div class="nav">
    <div class="nav-item">Home</div>
  </div>
</div>
<div class="main">
  <div class="article">
    <div class="title">Headline</div>
  </div>
</div>

<!-- ✅ Semantic HTML5 -->
<header>
  <nav>
    <a href="/">Home</a>
  </nav>
</header>
<main>
  <article>
    <h1>Headline</h1>
  </article>
</main>

Image aspect ratios

html
<!-- ❌ Distorted images -->
<img src="photo.jpg" width="300" height="100">
<!-- If actual ratio is 4:3, this squishes the image -->

<!-- ✅ Preserve aspect ratio -->
<img src="photo.jpg" width="300" height="225">
<!-- Actual 4:3 dimensions -->

<!-- ✅ CSS object-fit for flexibility -->
<img src="photo.jpg" style="width: 300px; height: 200px; object-fit: cover;">

Permissions & privacy

Request permissions properly

javascript
// ❌ Request on page load (bad UX, often denied)
navigator.geolocation.getCurrentPosition(success, error);

// ✅ Request in context, after user action
findNearbyButton.addEventListener('click', async () => {
  // Explain why you need it
  if (await showPermissionExplanation()) {
    navigator.geolocation.getCurrentPosition(success, error);
  }
});

Permissions policy

html
<!-- Restrict powerful features -->
<meta http-equiv="Permissions-Policy" 
      content="geolocation=(), camera=(), microphone=()">

<!-- Or allow for specific origins -->
<meta http-equiv="Permissions-Policy" 
      content="geolocation=(self 'https://maps.example.com')">

Audit checklist

Security (critical)

  • HTTPS enabled, no mixed content
  • No vulnerable dependencies (npm audit)
  • CSP headers configured (with frame-ancestors, base-uri, form-action)
  • require-trusted-types-for 'script' enforced (or report-only during rollout)
  • Third-party <script>/<link rel="stylesheet"> pinned with SRI hashes
  • Security headers present (HSTS, X-Content-Type-Options, Referrer-Policy)
  • No exposed source maps (and sourcesContent stripped from uploaded ones)

Compatibility

  • Valid HTML5 doctype
  • Charset declared first in head
  • Viewport meta tag present
  • No deprecated APIs used
  • Passive event listeners for scroll/touch

Code quality

  • No console errors
  • Valid HTML (no duplicate IDs)
  • Semantic HTML elements used
  • Proper error handling
  • Memory cleanup in components

UX

  • No intrusive interstitials
  • Permission requests in context
  • Clear error messages
  • Appropriate image aspect ratios

Tools

ToolPurpose
npm auditDependency vulnerabilities
SecurityHeaders.comHeader analysis
W3C ValidatorHTML validation
Live Lighthouse audit (Chrome DevTools MCP: lighthouse_audit)Rendered Best Practices checks for agents
Lighthouse CLIBest Practices audit fallback
ObservatorySecurity scan

References

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 Best Practices AI skill do?

Apply modern web development best practices for security, compatibility, and code quality. Use when asked to "apply best practices", "security audit", "modernize code", "code quality review", or "check for vulnerabilities".

Why use Best Practices on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/addyosmani/web-quality-skills/tree/main/skills/best-practices. 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 Best Practices?

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 Best Practices?

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

Is the Best Practices AI skill free?

Yes. It is published on GitHub by addyosmani 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 👇