Search Audit

Audit a single URL across four dimensions: SEO fundamentals, content structure, AI/GEO readiness, and E-E-A-T authority signals. Produces a scored report (out of 38) with prioritized fixes.

Usage

Ask your AI tool to audit a URL for search and AI readiness. You can optionally request a focus area — seo-only , geo-only , eeat-only , or content-only .

Process

Step 1: Fetch & Parse

Fetch the page at the provided URL. Extract:

• Full HTML source (for schema, meta tags, heading structure)

• Rendered page content (for content evaluation)

Then fetch these three files from the same domain:

• {domain}/robots.txt — check for AI crawler directives

• {domain}/sitemap.xml — check for existence

• {domain}/llms.txt — check for existence

Step 2: SEO Fundamentals (10 checks)

Evaluate traditional SEO hygiene. These are prerequisites — if broken, nothing else matters.

• HTTPS — Page is served over HTTPS

• Title tag — Exists, 50-60 characters, descriptive

• Meta description — Exists, 150-160 characters, includes target intent

• Single H1 — Exactly one H1 tag present

• Heading hierarchy — Logical order (H1 > H2 > H3), no skipped levels, no empty headings

• Image alt text — All images have descriptive alt attributes

• Clean URL — Descriptive, readable, no excessive parameters

• Internal links — Links to and from related pages on the site

• Crawlable content — Key content in standard HTML, not hidden behind JavaScript rendering

• Page speed — No obvious performance issues (heavy uncompressed images, render-blocking scripts, excessive DOM size)

Step 3: Content Structure (10 checks)

Evaluate how well the content is structured for both human readers and AI extraction.

• Direct answer up front — Each section leads with the key insight in the first 1-2 sentences

• Question-based headings — H2s/H3s mirror how users phrase queries (What is..., How do I..., Why does...)

• Short paragraphs — 2-4 sentences per paragraph, no wall-of-text blocks

• FAQ section — Dedicated Q&A section with real customer/user questions, 1-3 sentence answers

• Comparison tables — Side-by-side content uses tables (35% higher AI extractability)

• Lists for structure — Numbered lists for processes, bullet points for features/options

• "[TERM] is..." definitions — Key concepts defined with at least two defining sentences

• Conversational keywords — 5+ natural, long-tail query variations woven into content

• Intent clusters — Covers related follow-up questions a user would likely ask next

• Multimedia — Images, video, or infographics present as content signals

Step 4: AI & GEO Readiness (10 checks)

Evaluate technical signals that determine whether AI systems can find, extract, and cite this page.

• AI crawler access — robots.txt does not block GPTBot, ClaudeBot, or other AI user agents

• llms.txt exists — Domain has an llms.txt file communicating AI crawling policies

• Organization schema — JSON-LD with @type Organization present (site-wide entity definition)

• Article/Page schema — Appropriate JSON-LD for the content type (Article, HowTo, Product) with dateModified

• FAQPage schema — Separate FAQPage JSON-LD if FAQ section exists

• Author schema — Author entity linked in structured data

• BreadcrumbList schema — JSON-LD clarifying site hierarchy

• Schema validates — All structured data passes validation (well-formed JSON, no errors)

• Mobile responsive — Content readable and functional across screen sizes

• Sitemap inclusion — Domain has a sitemap.xml and it's accessible

Step 5: E-E-A-T & Authority (8 checks)

Evaluate trust signals that determine whether AI systems choose to cite this page over competitors.

• Author identified — Content has a named author, not "admin" or anonymous

• Author bio present — Author bio with credentials, expertise, and link to author page

• Author page exists — Dedicated author page with full credentials, external profiles, and publication history

• Sources cited — Key claims backed by specific data points with clear attribution

• Original data or insight — Contains first-party research, proprietary analysis, or unique expert perspective

• Freshness signals — Publication date and/or last-updated date visible on page

• Trust pages accessible — Privacy policy, terms, and relevant compliance pages exist and are linked

• About page with entity info — About Us/About page clearly defines the business entity

Output Format

Present the report in this exact structure:

Page Audit Report

URL: [url] Date: [current date] Overall Score: X/38

SEO Fundamentals (X/10)

✓/✗ HTTPS ✓/✗ Title tag — [note: actual title, character count] ✓/✗ Meta description — [note: actual description, character count] ✓/✗ Single H1 ✓/✗ Heading hierarchy ✓/✗ Image alt text — [note: X of Y images have alt text] ✓/✗ Clean URL ✓/✗ Internal links ✓/✗ Crawlable content ✓/✗ Page speed

Content Structure (X/10)

✓/✗ Direct answer up front ✓/✗ Question-based headings ✓/✗ Short paragraphs ✓/✗ FAQ section ✓/✗ Comparison tables ✓/✗ Lists for structure ✓/✗ "[TERM] is..." definitions ✓/✗ Conversational keywords ✓/✗ Intent clusters ✓/✗ Multimedia

AI & GEO Readiness (X/10)

✓/✗ AI crawler access — [note: which crawlers blocked/allowed] ✓/✗ llms.txt ✓/✗ Organization schema ✓/✗ Article/Page schema — [note: type found, dateModified present?] ✓/✗ FAQPage schema ✓/✗ Author schema ✓/✗ BreadcrumbList schema ✓/✗ Schema validates ✓/✗ Mobile responsive ✓/✗ Sitemap inclusion

E-E-A-T & Authority (X/8)

✓/✗ Author identified — [note: author name if found] ✓/✗ Author bio present ✓/✗ Author page exists ✓/✗ Sources cited — [note: X citations found] ✓/✗ Original data or insight ✓/✗ Freshness signals — [note: dates found] ✓/✗ Trust pages accessible ✓/✗ About page with entity info

Score Summary

Rating scale: 90%+ Excellent | 75-89% Good | 60-74% Needs Work | Below 60% Poor

Priority Fixes

[Top 5 failed criteria ranked by impact. For each:]

1. [Failed criterion] — [Why it matters + specific fix with actual suggested text/code where possible]
2. ...

Rules

• Be objective. If borderline, lean toward FAIL and explain in priority fixes.

• Be specific in fixes — suggest actual meta description text, actual schema JSON, actual heading rewrites. "Add a meta description" is not useful. "Add meta description: '[suggested text]'" is.

• For schema checks, search page source for <script type="application/ld+json"> and evaluate each block.

• Priority fixes should weight: technical foundation > AI readiness > content structure > E-E-A-T (you can't build authority on a broken page).

• If user specifies a focus area, still run all checks but only expand detail on the requested section.