Rapport du site : 1 page mesurée. Note globale : Faible.
Score du site · graza.co
Combine l'origine et 1 page mesurée. Le rapport ci-dessous détaille uniquement la page d'origine.
Transmettez les constats à ceux qui peuvent les résoudre. Copiez un résumé prêt pour Slack, Teams ou votre prochaine réunion.
Ce rapport est public. Votre équipe peut l’ouvrir sans compte.
Y a-t-il de la substance qu'un agent puisse lire ?
Homepage does not serve Markdown via content negotiation
Serve a Markdown representation of your pages when agents request "Accept: text/markdown". Agents like Claude Code and Cursor ask for it, and Markdown cuts token usage by roughly 80% against HTML. Cloudflare ("Markdown for Agents") and Vercel can enable this without code changes.
No <link rel="alternate" type="text/markdown"> fallback found on the homepage
If you cannot enable content negotiation, advertise a Markdown version with <link rel="alternate" type="text/markdown" href="/index.md"> so agents can discover it.
No JSON-LD structured data found
Add a <script type="application/ld+json"> block in your HTML <head> with schema.org structured data describing your site, organization, or person.
Sparse server-rendered content (13 words, 88 chars)
Render at least the main page content server-side. Many AI crawlers (GPTBot, ClaudeBot, CCBot) do not execute JavaScript and will see only the static HTML.
Text-to-markup ratio is healthy (17.0% of structural markup)
No semantic HTML landmarks found
Replace generic <div> structures with semantic tags: <main>, <article>, <header>, <nav>, <footer>. Agents use these to identify the primary content region.
Single <h1> heading: "Your connection needs to be verified before you can proceed"
<title> length 28 chars: "Verifying your connection..."
<meta name="description"> is missing
Add <meta name="description" content="..."> in <head> with a 70-160 character summary. Agents use this as the canonical short description.
No <link rel="canonical"> found
Add <link rel="canonical" href="https://your-site.com/page"> so agents have an unambiguous URL to cite even when crawled via a redirect or query-string variant.
<html lang="en">
UTF-8 charset declared
Viewport meta present: "width=device-width, initial-scale=1.0"
1 obstacle(s) on the entry page
A CAPTCHA or a meta refresh on the landing page stops an agent before it reaches any content. If bot protection is needed, apply it to the actions that need it, not to reading a page.
Method note: this reads markup, not a rendered accessibility tree
Un agent peut-il seulement la récupérer ?
Baseline homepage request failed — cannot compare crawler access
Homepage request failed — cannot assess crawl efficiency
A missing page redirects (301) instead of returning 404
A redirect on a nonexistent path hides the error. Return 404 or 410 so a client can tell the difference.
Homepage answers after 1 redirect
Rate limited with no Retry-After header
A 429 without Retry-After tells a well-behaved crawler nothing about when to return, so it either gives up or keeps hammering. Always send the header.
Content-Type: text/html with charset
Site is served over HTTPS
Could not verify HTTP→HTTPS redirect
Test manually: a request to http://your-site.com should respond with 301 → https://your-site.com.
No Strict-Transport-Security header
Add: Strict-Transport-Security: max-age=31536000; includeSubDomains; preload. This locks browsers and many agents to HTTPS for 1 year.
Homepage is indexable
No directive restricts how AI assistants may use this page
Un agent trouve-t-il ce que vous publiez ?
/llms.txt not found
Create a /llms.txt file at your site root following the llmstxt.org specification. It should be a Markdown file starting with "# Your Site Name" and include a description, sections, and links.
/robots.txt not found
Create a /robots.txt file at your site root. Add User-agent entries for AI crawlers (GPTBot, ClaudeBot, etc.) with Allow: / to grant access.
No sitemap found
Publish an XML sitemap at /sitemap.xml and reference it from robots.txt with: Sitemap: https://your-site.com/sitemap.xml
No AI meta tags (ai:*) found
Add AI meta tags to your HTML <head>: <meta name="ai:summary" content="Brief description">, <meta name="ai:content_type" content="website">, <meta name="ai:author" content="Your Name">.
No rel="alternate" link to llms.txt in HTML
Add to your <head>: <link rel="alternate" type="text/plain" href="/llms.txt" title="LLM-optimized content">
No rel="alternate" link to the Agent Card in HTML
Add to your <head>: <link rel="alternate" type="application/json" href="/.well-known/agent-card.json" title="Agent Card">
No rel="me" identity links found
Add rel="me" links to verify your identity across platforms: <link rel="me" href="https://github.com/yourname">, <link rel="me" href="https://twitter.com/yourname">.
No OpenGraph meta tags found
Add at minimum og:title, og:description, og:url, og:type, and og:image. Agents and link previews depend on these.
No Twitter Card meta tags found
Add twitter:card, twitter:title, twitter:description, and twitter:image so X / Threads / Bluesky / Discord agents render link previews correctly.
Missing critical header: Strict-Transport-Security
Add the Strict-Transport-Security response header to your server configuration. This is a critical security header.
5/7 security headers present
No Link header for AI discovery (llms.txt, Agent Card)
Add a Link response header pointing to your AI discovery files: Link: </llms.txt>; rel="alternate"; type="text/plain", </.well-known/agent-card.json>; rel="alternate"; type="application/json"
No machine-readable discovery relations beyond llms.txt and the Agent Card
Advertise what you publish with Link relations so agents stop guessing paths. Add the ones that apply, for example: Link: </llms.txt>; rel="describedby", </.well-known/api-catalog>; rel="api-catalog". Informational in 3.x: this does not affect your score.
Qu'un agent peut-il appeler ?
No agent resource catalog found
A catalog is one document listing everything an agent can call here — agent cards, MCP servers, APIs, skills — so a client stops probing four conventions to find out. Worth publishing once you have more than one of those. Informational: both ai-catalog.json and ard.json are still drafts, so this never affects your score.
Protected-resource metadata published (RFC 9728)
2 authorization server(s) named
Authorization server https://account.graza.co publishes no discovery metadata
The chain stops here: an agent knows which server to ask but not which endpoints it has. Publish RFC 8414 or OpenID discovery metadata.
UCP profile published at /.well-known/ucp
UCP version 2026-08-25
1 commerce service(s) declared: dev.ucp.shopping
2 declared schema URL(s) resolve
UCP profile declares no payment handlers
Without a payment handler an agent can read your catalog but cannot complete a purchase. Declare the handlers you accept.
UCP profile declares no signing keys
Publish the public keys an agent uses to verify your responses. Money is moving; identity should not rest on DNS alone.
No agent-facing surface — an Agent Card does not apply to this site
No API surface — API discovery does not apply to this site
No MCP server — MCP discovery does not apply to this site
No developer-facing surface — skills do not apply to this site
No forms and no WebMCP code — nothing here for an agent to invoke as a tool
Quels droits d'usage sont déclarés ?
/.well-known/security.txt not found
Create a /.well-known/security.txt file per RFC 9116. At minimum, include Contact: and Expires: fields. See https://securitytxt.org/ for a generator.
No RSL license discovery found
Declare machine-readable licensing terms for your content with Really Simple Licensing. Add to robots.txt: License: https://your-site.com/license.xml — then publish the RSL document. See https://rslstandard.org.
No machine-readable usage policy declared
State your terms where they can be read without a lawyer. The lowest-effort option is a Content-Signal line in robots.txt: Content-Signal: search=yes, ai-input=yes, ai-train=no. Absence is neutral, not permission — but it also gives you nothing to point at.
Offrez à votre équipe des rapports complets, des corrections prioritaires et un suivi pour préserver les progrès après chaque livraison.