What the free check actually returns.
A worked example on a real public site, using the same nine signals as the free check. It shows the format and depth of the evidence. Every row below records only what the fetched responses contained — nothing is inferred or invented.
Snapshot: 20 Nov 2026, UTC.
Method: the public homepage and four conventional paths were fetched once with a desktop user agent. A 404 at a conventional path is an observation, not proof a site cannot be read — some sites use different filenames or restrict crawlers deliberately. Findings can change; this is a neutral snapshot, not a comment on the product.
Result: 3 of 9 signals in place
One further signal (Open Graph) is partial: two of the three required tags are present. Partial signals do not count towards the score.
| Signal | Result | Evidence from the live responses |
|---|---|---|
/llms.txt with real content | Gap | GET /llms.txt returned HTTP 404; /llms-full.txt also 404. |
Named AI-crawler rules in robots.txt | Gap | /robots.txt returned HTTP 200 (1,247 B) with a content-signals notice, but no User-agent rules for GPTBot, ClaudeBot or PerplexityBot and no Sitemap: line. |
| Organisation and offering structured data | Gap | The homepage HTML (97,541 B) contains no application/ld+json block. |
| Descriptive title and meta-description | In place | <title>HTTPie – API testing client that flows with you</title>; one <meta name="description">. |
| Exactly one H1 | In place | One <h1> element in the homepage HTML; its text is “FLOW”. |
Reachable XML sitemap linked from robots.txt | Gap | /sitemap.xml, /sitemap_index.xml and /sitemap-index.xml all returned 404, and robots.txt has no Sitemap: line. |
| One canonical URL per page | Gap | No rel="canonical" in the homepage HTML. |
| Open Graph tags | Partial | og:title and og:description present; og:url absent. The check requires all three. |
| Useful text in the HTML, not only JavaScript | In place | The main copy (“API testing client”) appears directly in the returned HTML. |
The three gaps I would fix first
- Publish
/llms.txt. A short markdown index of the pages you want assistants to read and quote. Cheap, static, no dependency. - Add a real sitemap and crawler rules. Generate the XML sitemap, declare it with a
Sitemap:line inrobots.txt, and add explicitUser-agentrules for the AI crawlers you want to allow. - Add the missing template tags.
OrganizationplusSoftwareApplicationJSON-LD, arel="canonical", andog:url— three small edits that apply site-wide from one template.
Get the same evidence for your site
The free check verifies these nine signals against your live pages and returns the top three measurable gaps with the evidence, by email, with no obligation. Request a free check → Or run the two-minute self-check first; it stays in your browser and sends nothing.
The €19 Audit Pack adds the ranked fix report, draft llms.txt, structured-data block and a before/after re-check. The €49 tier adds a small tested MCP server for your public docs and FAQ. See the offer →
Ready to order? Order the €19 pack → · Order the €49 pack + MCP server →