our-claude-skills/custom-skills/_archive/seo-audit-agent/desktop-skill-refer/SEO_AUDIT_KNOWLEDGE.md

OurDigital SEO Audit - Claude Desktop Project Knowledge

Overview

This project knowledge file enables Claude Desktop to perform comprehensive SEO audits using MCP tools. It provides workflows for technical SEO analysis, schema validation, sitemap checking, and Core Web Vitals assessment.

Available MCP Tools

Primary Tools

Tool	Purpose
firecrawl	Website crawling, scraping, structured data extraction
perplexity	AI-powered research, competitive analysis
notion	Store audit findings in database
fetch	Fetch web pages and resources
sequential-thinking	Complex multi-step analysis

Tool Usage Patterns

Firecrawl - Web Scraping

firecrawl_scrape: Scrape single page content
firecrawl_crawl: Crawl entire website
firecrawl_extract: Extract structured data
firecrawl_map: Get site structure

Notion - Output Storage

API-post-search: Find existing databases
API-post-database-query: Query database
API-post-page: Create finding pages
API-patch-page: Update findings

SEO Audit Workflows

1. Full Site Audit

User prompt: "Perform SEO audit for https://example.com"

Workflow:

1. Use firecrawl_scrape to get homepage content
2. Use firecrawl_map to discover site structure
3. Check robots.txt at /robots.txt
4. Check sitemap at /sitemap.xml
5. Extract and validate schema markup
6. Use perplexity for competitive insights
7. Store findings in Notion database

2. Schema Markup Validation

User prompt: "Validate schema on https://example.com"

Workflow:

1. Use firecrawl_scrape with extractSchema option
2. Look for JSON-LD in <script type="application/ld+json">
3. Check for Microdata (itemscope, itemtype, itemprop)
4. Validate against schema.org requirements
5. Check Rich Results eligibility

Schema Types to Check:

• Organization / LocalBusiness
• Product / Offer
• Article / BlogPosting
• FAQPage / HowTo
• BreadcrumbList
• WebSite / WebPage

Required Properties by Type:

Type	Required	Recommended
Organization	name, url	logo, sameAs, contactPoint
LocalBusiness	name, address	telephone, openingHours, geo
Product	name	image, description, offers, brand
Article	headline, author, datePublished	image, dateModified, publisher
FAQPage	mainEntity (Question + Answer)	-

3. Robots.txt Analysis

User prompt: "Check robots.txt for example.com"

Workflow:

1. Fetch https://example.com/robots.txt
2. Parse directives:
   • User-agent rules
   • Disallow patterns
   • Allow patterns
   • Crawl-delay
   • Sitemap declarations
3. Check for issues:
   • Blocking CSS/JS resources
   • Missing sitemap reference
   • Overly restrictive rules

Sample Analysis Output:

Robots.txt Analysis
==================
User-agents: 3 defined (*, Googlebot, Bingbot)

Directives:
- Disallow: /admin/, /private/, /tmp/
- Allow: /public/, /blog/
- Sitemap: https://example.com/sitemap.xml

Issues Found:
- WARNING: CSS/JS files may be blocked (/assets/)
- OK: Sitemap is declared
- INFO: Crawl-delay set to 10s

4. Sitemap Validation

User prompt: "Validate sitemap at https://example.com/sitemap.xml"

Workflow:

1. Fetch sitemap XML
2. Parse and validate structure
3. Check:
   • XML syntax validity
   • URL count (max 50,000)
   • Lastmod date formats
   • URL accessibility (sample)
4. For sitemap index, check child sitemaps

Validation Criteria:

• Valid XML syntax
• <urlset> or <sitemapindex> root element
• Each <url> has <loc> element
• <lastmod> in W3C datetime format
• File size under 50MB uncompressed

5. Core Web Vitals Check

User prompt: "Check Core Web Vitals for https://example.com"

Workflow:

1. Use PageSpeed Insights (if API available)
2. Or analyze page with firecrawl for common issues
3. Check for:
   • Large images without optimization
   • Render-blocking resources
   • Layout shift causes
   • JavaScript execution time

Metrics & Thresholds:

Metric	Good	Needs Improvement	Poor
LCP	< 2.5s	2.5s - 4.0s	> 4.0s
CLS	< 0.1	0.1 - 0.25	> 0.25
FID/INP	< 100ms/200ms	100-300ms/200-500ms	> 300ms/500ms

6. Technical SEO Check

User prompt: "Check technical SEO for https://example.com"

Workflow:

1. Check HTTPS implementation
2. Verify canonical tags
3. Check meta robots tags
4. Analyze heading structure (H1-H6)
5. Check image alt attributes
6. Verify Open Graph / Twitter Cards
7. Check mobile-friendliness indicators

Checklist:

• HTTPS enabled
• Single canonical URL per page
• Proper robots meta tags
• One H1 per page
• All images have alt text
• OG tags present (og:title, og:description, og:image)
• Twitter Card tags present
• Viewport meta tag for mobile

7. Local SEO Audit

User prompt: "Local SEO audit for [Business Name] in [Location]"

Workflow:

1. Search for business citations with perplexity
2. Check for LocalBusiness schema
3. Verify NAP (Name, Address, Phone) consistency
4. Look for review signals
5. Check Google Business Profile (manual)

Notion Database Integration

Default Database

• Database ID: 2c8581e5-8a1e-8035-880b-e38cefc2f3ef
• Name: OurDigital SEO Audit Log

Database Properties

Property	Type	Values
Issue	Title	Finding title
Site	URL	Audited site URL
Category	Select	Technical SEO, On-page SEO, Content, Local SEO, Performance, Schema/Structured Data, Sitemap, Robots.txt
Priority	Select	Critical, High, Medium, Low
Status	Status	Not started, In progress, Done
URL	URL	Specific page with issue
Found Date	Date	Discovery date
Audit ID	Rich Text	Groups findings from same audit

Page Content Template

Each finding page should contain:

## Description
[Detailed explanation of the issue]

## Impact
[Business/ranking impact with warning callout]

## Recommendation
[Actionable solution with lightbulb callout]

Creating Findings

Use Notion MCP to create pages:

1. Query database to check for existing entries
2. Create new page with properties
3. Add content blocks (Description, Impact, Recommendation)

Schema Markup Templates

Organization Schema

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "[Company Name]",
  "url": "[Website URL]",
  "logo": "[Logo URL]",
  "sameAs": [
    "[Social Media URLs]"
  ],
  "contactPoint": {
    "@type": "ContactPoint",
    "telephone": "[Phone]",
    "contactType": "customer service"
  }
}

LocalBusiness Schema

{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "[Business Name]",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "[Street]",
    "addressLocality": "[City]",
    "addressCountry": "[Country Code]"
  },
  "telephone": "[Phone]",
  "openingHoursSpecification": [{
    "@type": "OpeningHoursSpecification",
    "dayOfWeek": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"],
    "opens": "09:00",
    "closes": "18:00"
  }]
}

Article Schema

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "[Article Title]",
  "author": {
    "@type": "Person",
    "name": "[Author Name]"
  },
  "datePublished": "[ISO Date]",
  "dateModified": "[ISO Date]",
  "publisher": {
    "@type": "Organization",
    "name": "[Publisher Name]",
    "logo": {
      "@type": "ImageObject",
      "url": "[Logo URL]"
    }
  }
}

FAQPage Schema

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "[Question Text]",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "[Answer Text]"
    }
  }]
}

BreadcrumbList Schema

{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [{
    "@type": "ListItem",
    "position": 1,
    "name": "Home",
    "item": "[Homepage URL]"
  }, {
    "@type": "ListItem",
    "position": 2,
    "name": "[Category]",
    "item": "[Category URL]"
  }]
}

Common SEO Issues Reference

Critical Priority

• Site not accessible (5xx errors)
• Robots.txt blocking entire site
• No HTTPS implementation
• Duplicate content across domain
• Sitemap returning errors

High Priority

• Missing or duplicate title tags
• No meta descriptions
• Schema markup errors
• Broken internal links
• Missing canonical tags
• Core Web Vitals failing

Medium Priority

• Missing alt text on images
• Thin content pages
• Missing Open Graph tags
• Suboptimal heading structure
• Missing breadcrumb schema

Low Priority

• Missing Twitter Card tags
• Suboptimal URL structure
• Missing FAQ schema
• Review schema not implemented

Quick Commands Reference

Task	Prompt
Full audit	"Perform SEO audit for [URL]"
Schema check	"Validate schema on [URL]"
Sitemap check	"Validate sitemap at [URL]"
Robots.txt	"Analyze robots.txt for [domain]"
Performance	"Check Core Web Vitals for [URL]"
Generate schema	"Generate [type] schema for [details]"
Export to Notion	"Export findings to Notion"
Local SEO	"Local SEO audit for [business] in [location]"
Competitive	"Compare SEO of [URL1] vs [URL2]"

Tips for Best Results

1. Be specific - Provide full URLs including https://
2. One site at a time - Audit one domain per session for clarity
3. Check Notion - Review existing findings before creating duplicates
4. Prioritize fixes - Focus on Critical/High issues first
5. Validate changes - Re-audit after implementing fixes

Limitations

• No direct Python script execution (use MCP tools instead)
• PageSpeed API requires separate configuration
• Google Search Console data requires authenticated access
• GA4 data requires service account setup
• Large sites may require multiple sessions