What a VERIS implementation actually changes.
This walk-through uses a fictional business so the logic can be shown clearly without exposing a real client. The proof pattern is the important part: before, implementation, and after, all tied to public checks.
Fictional boutique hotel used purely for illustration. The point is to show the type of proof package VERIS produces, not to imply a published client case.
Three structural changes, each with a visible proof path.
Schema moved from absent to valid.
The business became machine-readable as a LodgingBusiness entity with the required structure present.
Crawler access moved from blocked to explicit allow.
The site stopped preventing relevant published AI/search agents from reaching the pages.
llms.txt moved from missing to live.
The business gained a cleaner machine-readable summary that can be checked directly at the URL.
Schema markup, before and after.
Before the implementation, Magnolia Inn had no usable JSON-LD schema on the page. After implementation, Rich Results Test can detect a valid LodgingBusiness entity with the expected structure present.
Schema markup validation
robots.txt, from broad blocking to explicit crawl access.
Before the implementation, the wildcard block prevented relevant agents from reaching the site. After the update, the directives make the crawl path clear and publish the sitemap in the same file.
# Before VERIS Implementation User-agent: * Disallow: / # OpenAI and Perplexity crawlers blocked by wildcard # Critical finding - all crawlers effectively blocked
# After VERIS Implementation User-agent: * Allow: / User-agent: OAI-SearchBot Allow: / User-agent: GPTBot Allow: / User-agent: PerplexityBot Allow: / User-agent: Google-Extended Allow: / User-agent: ClaudeBot Allow: / Sitemap: https://magnoliainn.com/sitemap.xml
llms.txt, from missing to deployed.
Before the implementation, the file returned a 404. After deployment, the business has a published AI-readable summary that can be checked directly at the URL.
llms.txt deployment
The proof is not the code change alone. It is the evidence bundle around it.
A verified implementation is delivered with the supporting outputs that show the change clearly enough for the client to review without relying on blind trust.
- vImplemented files live on the domain
- vRich Results Test outputs before and after
- vSchema validator output after implementation
- vrobots.txt before and after showing relevant directives
- vllms.txt live check at the published URL
- vPublic discovery screenshots where relevant
- vA delivery note tying each proof item back to the work completed
Need the methodology behind these checks?
The verification tools page explains where each proof item comes from, and the audit applies the same logic to your own site before implementation begins.