Elementera
Products
OriginumLenzThe Supervised plan, our managed AEO service
Case StudiesBlogFree ToolsResearchAbout UsBook a call
Book a call
  1. Home
  2. Free tools
  3. AI search technical checklist

AI search technical checklist

Key point. These are the 30 technical checks our client audits run before any change goes live: the technical baseline, AI crawler access and llms.txt, and accessibility and agent readiness. Tick what is done; your ticks stay in your browser, and a share link carries them to a colleague.

AI search technical checklist

30 checks from our client audit procedures: the technical baseline, AI crawler access and llms.txt, and accessibility and agent readiness. Tick what is done. The severity labels are our suggested order, not a verdict about your site.

Technical baseline
AI crawler access and llms.txt
Accessibility and agent readiness

Your ticks stay in this browser. A share link carries them in the address after the # sign, which browsers do not send to any server.

What this did not check. It does not test your site: every tick is your own reading. Each line is a check our client audits run, and a passing line still needs its evidence written down. Next: AI crawler check. There is nothing to sample: start ticking.

Where do the checks come from?

From the checklists at the end of our own client procedures for a technical baseline audit, AI crawler access and llms.txt, and accessibility and agent readiness. The page is built from those procedures, so when a check is added there it appears here at the next build.

What do the severity labels mean?

They are our suggested order. High covers anything that can stop a crawler or an AI fetcher reading a page: robots rules, noindex, canonicals, broken links and redirects, content that needs JavaScript. Low covers process checks, such as keeping a change log. Medium is everything else. Your own site may need a different order.

How does the technical checklist work?

The AI search technical checklist holds 30 checks from our client procedures. They are read at build time from each procedure's 'Checks before going live' lines, so a check added there appears here at the next build. Each gets a suggested severity by one fixed rule: 12 High (anything that can stop a crawler or an AI fetcher reading a page), 12 Medium and 6 Low (process checks). Built by Elementera. Last checked on November 3, 2026, against those procedures.

Worked example: one High check, found and fixed

For example, one High check found and fixed: a site's robots.txt has a group for OAI-SearchBot with Disallow: /blog/. The check "Every search crawler and user fetcher allowed for home, article and product paths" stays unticked: ChatGPT search cannot read any article. The fix is to remove that line, then paste the whole file into the AI crawler check, which reads the home, article and product paths for each crawler. Tick the line only when the check shows every search crawler allowed, and write down the date and the result as its evidence.

The 12 High checks

The technical checklist marks 12 of its 30 checks High: fix these first.

SectionCheck
Technical baselineSearch-parameter URLs: probes carry X-Robots-Tag: noindex, controls carry none (step 21b)
Technical baseline0 pages without canonical, <main>, H1, og:image, unique description (or each exception named)
Technical baseline0 internal links to a 404 or a redirect
Technical baselineEvery navbar, footer and side-panel link answers 200 directly; every required hub is linked from the chrome and listed in the sitemap
Technical baselineShare previews per template read from the server HTML: og:title and og:description say what the page's own title and description say (no stale or tool-made copy) and og:url equals the canonical; og:image is an absolute URL that answers 200 and is at least 1200 x 630; twitter:card is summary_large_image with an image
Technical baselineOn-page completeness read from the server HTML: no page missing a title, description, canonical, lang, Open Graph, Twitter card, JSON-LD, icon or usable image alt (exceptions listed with a reason)
AI crawler access and llms.txtEvery search crawler and user fetcher allowed for home, article and product paths
AI crawler access and llms.txtCDN/WAF: 0 actions on verified AI search bots in the window; impostors reported separately
AI crawler access and llms.txtEach bot has one verdict from verified requests only; "not observed" is never written as "blocked"
AI crawler access and llms.txtMain content readable with JavaScript off
AI crawler access and llms.txtllms.txt links unique, internal ones in the sitemap and 200 without a redirect, no dead external link
AI crawler access and llms.txtWhere the site sits behind Cloudflare: every AI operator the client allows shows no 403 or 404 in AI Operator

FAQ

Is anything I tick sent to Elementera?

No. Ticks are saved in your browser. A share link puts them in the address after the # sign, which browsers do not send to any server.

Does a full checklist mean AI engines will cite my site?

No. It means nothing technical is in the way. Whether an engine names you depends on what your pages say and who else covers the topic.

Work with us

Want to see whether ChatGPT and Gemini name your company for your buyers' questions? Get your free AI visibility report from Lenz. Want the checks run for you? Book a call about managed AEO services.

Elementera
Expertise in. Original content out.

Elementera turns what a business knows into original content for AI search, and measures how AI engines describe and recommend it.

Product
OriginumManaged AEO ServicesLenzBook a Call
Learn
BlogCase StudiesFree ToolsResourcesResearchEvents
Guides
What is AEOAEO vs SEOAI visibility funnelllms.txt guide
Company
About UsContactSite map
hello@elementera.comTrustPrivacy PolicyTerms of Service

Supported by

Google
Microsoft
NVIDIA
Cloudflare
Anthropic

Machine-readable summary: llms.txt ยท Company facts