What the site check is
Give it a website and it reads the homepage and the files an engine looks for: robots.txt, the sitemap, llms.txt, the structured data, the page as it renders without JavaScript, and on the paid plans every page of the site. It runs 100 checks in about half a minute and reports each one: what it found, why it matters, and what to do. This reference is those 100 checks, written out.
The score is a ceiling, not a prediction. A site an engine cannot read cannot be recommended, so a failing check caps you. A clean score means nothing on your side is in the way. Whether an engine then names you is measured by asking the engines, which is the other half of the product.
The five groups
How a check is weighted
Each check carries an impact from critical to low, and a note on how directly it reaches AI search: 18 of the 100 affect it directly, the rest are what AI search inherits from ordinary search, or barely touch it. 18 checks are critical: the ones that stop an engine reading the site at all, such as a robots rule that blocks the retrieval crawlers or a page with no text until a script runs. A check that does not apply to a site, such as opening hours for a business with no premises, is not scored against it.
Impact and reach are stated beside every check on the group pages, so a reader can see at a glance which failing item to fix first.
All 100 checks
Being found by AI
- AI crawler access
- Search crawler access
- llms.txt
- llms-full.txt
- ai.txt
- TDM reservation
- Content signals
- AI crawler edge access
- Interior pages
- New URLs are real pages
- Citable surface
- Markdown twin
- Answer-shaped content
- Dates & authorship
- JS-free content
- Reached by a public crawler
- Knowledge graph entity
- Bing index
- IndexNow
Being found on Google
- robots.txt
- Sitemap directive
- sitemap.xml
- Sitemap contents
- Sitemap index
- Content feed
- www vs non-www
- <html lang>
- Schema completeness
- Heading structure
- Response time
- Redirect chain
- Trailing slash
- JavaScript dependence
- Mixed content
- Mobile rendering
- Page weight & blocking
- Image delivery
- On-page subject
- Core Web Vitals
- Critical path
- Layout stability
- Competing pages
- Caching & validators
- Entity assets crawlable
- Rating backed by page
- Listed URLs resolve
- Share image
- Canonical URL
- Title tag
- Meta description
- Meta robots
- Snippet & preview directives
- Open Graph tags
- hreflang
- Canonical target
- Viewport meta
- Title & description uniqueness
- Image alt text
- Page-level schema
- JSON-LD
- Organization schema
- WebSite schema
- BreadcrumbList schema
- FAQPage schema
- 404 status code
- HTTPS redirect
- Lab performance
Your location
- Local entity
- Address completeness
- PIN code & state
- Telephone format
- Address backed by the page
- Opening hours
- Coordinates & map
- Branch markup
- City pages
- Business Profile
- Listing agrees with the site
- Reviews
- Directory profiles
- PIN code in the directory
- Coordinates match the address
AI assistants doing things
General housekeeping
- Compression
- HSTS
- app-ads.txt
- security.txt
- Apple App Site Association
- Android asset links
- ads.txt
- favicon.ico
- favicon.svg
- apple-touch-icon.png
- Web manifest
- /.well-known/ served
- Legacy files
- Trackers and consent
Run it on your site
- Create a free account. No card.
- Enter the site. The score by group appears in about half a minute, with the name of every failing check.
- Fix what fails and run it again; the free plan has four checks a month. On the paid plans each failing check comes with the code to paste, written for your site and the platform it runs on, and an assistant in your site’s code can apply them through the MCP server.