Skip to content

rohandey/lighthouse-checker-skill

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Lighthouse Checker Skill

An AI skill for running Google Lighthouse audits across websites. Automatically discovers pages via sitemap.xml and generates organized HTML reports with pass/fail classification.

Features

  • Sitemap crawling - Automatically discovers pages from sitemap.xml
  • Single page mode - Check specific URLs directly
  • Category filtering - Run only Performance, Accessibility, Best Practices, or SEO
  • Pass/Fail organization - Reports sorted by score threshold
  • Interactive HTML dashboard - Summary with links to all reports
  • AI Skill - Works with Claude Code and other LLMs

Installation

Prerequisites

  • Node.js 14+ and npm
  • Python 3 (for score parsing)
  • Chrome/Chromium browser

For Claude Code

Clone the skill into your project's .claude/skills/ directory:

git clone https://github.com/rohandey/lighthouse-checker-skill.git .claude/skills/lighthouse-checker

Or add to your home directory for global access:

git clone https://github.com/rohandey/lighthouse-checker-skill.git ~/.claude/skills/lighthouse-checker

The skill is now available and will be triggered automatically when you use relevant prompts.

For Other LLMs (ChatGPT, Gemini, etc.)

Option 1: System Prompt Integration

Copy the contents of SKILL.md into your LLM's system prompt or custom instructions. This gives the LLM knowledge of the skill's capabilities and command syntax.

Option 2: Tool/Function Definition

Add as a tool definition in your LLM configuration:

{
  "name": "lighthouse_checker",
  "description": "Run Google Lighthouse audits on websites for performance, accessibility, best practices, and SEO",
  "parameters": {
    "url": "URL to audit (required)",
    "categories": "perf, a11y, bp, seo (optional, comma-separated)",
    "max_urls": "Maximum pages to crawl (optional, default: 150)",
    "threshold": "Pass threshold 0-100 (optional, default: 90)"
  }
}

Option 3: Local Setup

Clone the repository and reference it in your LLM's workspace:

git clone https://github.com/rohandey/lighthouse-checker-skill.git

Then instruct your LLM to use the script at scripts/check_lighthouse.sh when website auditing is requested.


AI Prompts

Once installed, you can use natural language prompts to trigger the skill. The skill will be automatically detected when you use these phrases.

Full Website Audit

Audit an entire website by crawling its sitemap:

Prompt Description
"Check page speed for https://example.com" Runs full Lighthouse audit on all pages found in sitemap
"Run a lighthouse audit on https://example.com" Same as above - audits entire site
"Audit website https://example.com" Comprehensive audit of all categories
"Test website performance for https://example.com" Full site audit with focus on performance
"Generate lighthouse report for https://example.com" Creates HTML reports for all discovered pages
"Analyze https://example.com for page speed issues" Identifies performance bottlenecks across the site
"Check Core Web Vitals for https://example.com" Audits LCP, FID, CLS metrics site-wide

Single Page Audit

Audit a specific page without crawling the sitemap:

Prompt Description
"Check lighthouse score for https://example.com/about" Audits only the /about page
"Audit this page: https://example.com/contact" Single page audit with all categories
"Run performance test on https://example.com/pricing" Tests one specific page
"Check https://example.com/blog/my-post for accessibility" Single page accessibility check
"How fast is https://example.com/products loading?" Performance audit for one page
"Test https://example.com/checkout page speed" Audit checkout page performance

Accessibility Audits

Focus specifically on accessibility (WCAG compliance):

Prompt Description
"Check accessibility for https://example.com" Full site accessibility audit
"Run a11y check on https://example.com" Same as above using shorthand
"Check WCAG compliance for https://example.com" Audits against WCAG guidelines
"Is https://example.com accessible?" Accessibility score and issues
"Find accessibility issues on https://example.com" Lists a11y violations
"Check screen reader compatibility for https://example.com" Focuses on assistive technology support
"Audit https://example.com for accessibility violations" Detailed a11y issue report
"Does https://example.com/signup meet accessibility standards?" Single page a11y check

Performance Audits

Focus specifically on speed and Core Web Vitals:

Prompt Description
"Test performance of https://example.com" Performance-only audit
"Check Core Web Vitals for https://example.com" LCP, FID, CLS metrics
"Analyze site speed for https://example.com" Detailed speed analysis
"Why is https://example.com slow?" Identifies performance bottlenecks
"Check load time for https://example.com" Time-to-interactive metrics
"Run performance audit on https://example.com" Full perf category audit
"How can I speed up https://example.com?" Performance recommendations
"Check render blocking resources on https://example.com" Specific perf issue

SEO Audits

Focus specifically on search engine optimization:

Prompt Description
"Check SEO score for https://example.com" SEO-only audit
"Audit SEO for https://example.com" Full SEO category check
"Is https://example.com SEO optimized?" SEO score and issues
"Check meta tags on https://example.com" SEO metadata audit
"Find SEO issues on https://example.com" Lists SEO violations
"Is https://example.com mobile friendly?" Mobile SEO check
"Check https://example.com crawlability" Search engine crawl audit
"How can I improve SEO for https://example.com?" SEO recommendations

Best Practices Audits

Focus on security, modern APIs, and code quality:

Prompt Description
"Check best practices for https://example.com" Best practices audit
"Run bp check on https://example.com" Same using shorthand
"Check security issues on https://example.com" Security-focused audit
"Does https://example.com use HTTPS?" Security check
"Find console errors on https://example.com" JavaScript error detection
"Check https://example.com for deprecated APIs" Modern standards audit
"Is https://example.com using modern web practices?" Full BP category

Multi-Category Audits

Combine multiple categories in one audit:

Prompt Description
"Check performance and SEO for https://example.com" Two categories
"Run accessibility and best practices audit on https://example.com" a11y + bp
"Test a11y and performance for https://example.com" Using shorthand
"Check perf, seo, and a11y for https://example.com" Three categories
"Audit https://example.com for performance and accessibility" perf + a11y
"Run SEO and best practices check on https://example.com" seo + bp

Advanced Options

Customize the audit with specific parameters:

Prompt Description
"Audit https://example.com with 80% pass threshold" Custom pass/fail threshold
"Check https://example.com, max 20 pages" Limit sitemap crawl
"Run lighthouse on https://example.com, save to ./my-reports" Custom output directory
"Audit https://example.com with 70% threshold and max 10 pages" Multiple options
"Check top 5 pages of https://example.com for performance" Limited crawl + category
"Quick audit of https://example.com (5 pages only)" Fast limited check
"Full deep audit of https://example.com (100 pages)" Extended crawl

Comparative & Follow-up Prompts

Prompt Description
"Re-run the lighthouse audit" Repeat last audit
"Check if https://example.com improved" Compare to previous run
"Show me the failed pages from the last audit" Review failures
"What pages failed accessibility?" Filter by category failure
"Open the lighthouse summary report" View generated dashboard
"Which pages have the lowest performance scores?" Analyze results

Script Usage

Basic Commands

# Check entire website (crawls sitemap.xml)
./scripts/check_lighthouse.sh -u https://example.com

# Check single page
./scripts/check_lighthouse.sh -u https://example.com/about

# Check accessibility only
./scripts/check_lighthouse.sh -u https://example.com -c a11y

# Check performance only
./scripts/check_lighthouse.sh -u https://example.com -c perf

Full Syntax

./scripts/check_lighthouse.sh -u <url> [-c <categories>] [-o <output_dir>] [-m <max_urls>] [-p <pass_threshold>]

Options:
  -u    URL to check (required)
        Domain (https://example.com) → crawls sitemap.xml
        Specific page (https://example.com/about) → checks only that page

  -c    Categories to check (comma-separated, default: all)
        performance (perf), accessibility (a11y), best-practices (bp), seo

  -o    Output directory (default: ./lighthouse-reports)

  -m    Maximum URLs from sitemap (default: 150)

  -p    Pass threshold 0-100 (default: 90)

Script Examples

# All categories, entire site
./scripts/check_lighthouse.sh -u https://mysite.com

# Accessibility only
./scripts/check_lighthouse.sh -u https://mysite.com -c accessibility

# Performance and SEO
./scripts/check_lighthouse.sh -u https://mysite.com -c perf,seo

# Single page, all categories
./scripts/check_lighthouse.sh -u https://mysite.com/contact

# Custom threshold and limit
./scripts/check_lighthouse.sh -u https://mysite.com -m 20 -p 80

Output

Reports are saved to a timestamped folder (e.g., ./lighthouse-reports-2024-01-15_14-30-22/):

lighthouse-reports-2024-01-15_14-30-22/
├── summary.html      # Dashboard with all results
├── pass/             # Pages scoring >= threshold
│   ├── page1.html
│   └── page1.score
└── fail/             # Pages scoring < threshold
    ├── page2.html
    └── page2.score

Each run creates a new timestamped folder, preserving previous reports for comparison.

Open summary.html in a browser to view all reports with scores for each category.


Categories

Category Short What It Measures
performance perf Core Web Vitals, load times, rendering
accessibility a11y WCAG compliance, screen reader support
best-practices bp Security, modern APIs, console errors
seo seo Meta tags, crawlability, mobile-friendly

Score Interpretation

Score Rating
90-100 Good (green)
50-89 Needs Improvement (orange)
0-49 Poor (red)

Project Structure

lighthouse-checker/
├── README.md                 # This file
├── SKILL.md                  # AI skill definition
├── scripts/
│   └── check_lighthouse.sh   # Main executable
├── references/
│   └── wcag-quick-ref.md     # WCAG accessibility guide
└── .claude/
    └── settings.local.json   # Claude Code settings

License

MIT

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages