Perf Audit logo

Perf Audit

Community
Houseofmvps
perf-audit

Run Lighthouse performance audit with auto-fix for common issues. Use when user wants to check or improve site performance.

Overview

PublisherHouseofmvps
Repositoryultraship
Skill nameperf-audit
Stars
122
Forks
14
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 Houseofmvps on GitHub. Read the source before you install it.

Installation

Install the Perf Audit 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/Houseofmvps/ultraship.git /tmp/ultraship
mkdir -p .claude/skills
cp -r /tmp/ultraship/skills/perf-audit .claude/skills/perf-audit
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Perf Audit 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 Perf Audit 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 Perf Audit 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.

Performance Audit

Run Lighthouse against the project and fix performance issues.

Process

Step 1: Find the URL

  • Check if dev server is running (try common ports: 3000, 5173, 4321, 8080)
  • Check package.json scripts for dev/start commands
  • Ask user for URL if not auto-detected
  • If no server running, suggest starting one first

Step 2: Run Lighthouse

bash
node ${CLAUDE_PLUGIN_ROOT}/tools/lighthouse-runner.mjs <url>

Parse JSON output for scores and opportunities.

Step 3: Report Scores & Core Web Vitals

Present all four Lighthouse scores:

  • Performance (target 90+)
  • Accessibility (target 90+)
  • Best Practices (target 90+)
  • SEO (target 100)

Report Core Web Vitals with targets:

  • LCP (Largest Contentful Paint): target <2.5s — report the LCP element from lcp_element
  • TBT (Total Blocking Time): target <200ms — proxy for INP
  • CLS (Cumulative Layout Shift): target <0.1
  • FCP (First Contentful Paint): target <1.8s
  • SI (Speed Index): target <3.4s

Step 4: Identify Root Causes

Before fixing, identify what's actually causing problems:

LCP too high? Check lcp_element in results — fix THAT specific element:

  • If it's an <img>: preload it, use WebP, add fetchpriority="high"
  • If it's text: check web font loading, add font-display: swap
  • If it's a background image: preload via <link rel="preload">

CLS too high? Check diagnostics for layout shift sources:

  • Images without width/height → add dimensions (scanner already checks this)
  • Dynamic content injected above fold → reserve space with CSS
  • Web fonts causing FOUT → add font-display: optional or swap

Unused resources? Check wasted_resources:

  • unused_js_kb > 100KB → code splitting or tree shaking needed
  • unused_css_kb > 50KB → purge unused CSS (PurgeCSS, Tailwind purge)

Third-party scripts slow? Check third_party_impact:

  • Analytics blocking >100ms → defer/async load
  • Ad scripts blocking >200ms → lazy load below fold
  • Chat widgets → load on user interaction, not page load

Step 5: Apply Fixes

For each opportunity, apply fixes using Edit tool:

Performance fixes (high impact first):

  • Render-blocking resources → add defer or async to non-critical scripts
  • Unused JavaScript → recommend code splitting, dynamic imports, tree shaking
  • Unused CSS → recommend PurgeCSS or Tailwind purge config
  • LCP image not preloaded → add <link rel="preload" as="image" href="...">
  • Images not lazy-loaded → add loading="lazy" to below-fold images
  • Images without dimensions → add width and height attributes
  • No preconnect → add <link rel="preconnect" href="..."> for external origins
  • Font display → add font-display: swap to @font-face declarations
  • Third-party scripts → defer non-essential scripts, lazy load on interaction

Accessibility fixes:

  • Missing alt text (also caught by SEO scanner)
  • Missing form labels → add <label> elements
  • Low contrast → adjust colors
  • Missing landmark regions → add <main>, <nav>, <footer>

Step 6: Graceful Degradation

  • No Chrome: report "Chrome needed for Lighthouse. Install Chrome or Chromium."
  • Lighthouse timeout: return partial results with warning
  • No dev server: suggest starting one, or test against production URL

Key Principle

Fix, don't just audit. Identify the specific elements causing problems. Apply every automated fix possible. For build-tool changes (code splitting, tree shaking), provide exact config recommendations.

Frequently asked questions

What does the Perf Audit AI skill do?

Run Lighthouse performance audit with auto-fix for common issues. Use when user wants to check or improve site performance.

Why use Perf Audit on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Houseofmvps/ultraship/tree/main/skills/perf-audit. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Perf Audit?

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 Perf Audit?

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

Is the Perf Audit AI skill free?

Yes. It is published on GitHub by Houseofmvps 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 👇