Website-Bericht: 1 gemessene Seite. Gesamtbewertung: Befriedigend.
Site-Score · notion.so
Verrechnet den Origin und 1 gemessene Seite. Der Bericht unten behandelt allein die Origin-Seite.
Bringe die Ergebnisse zu den Menschen, die sie beheben können. Kopiere eine fertige Zusammenfassung für Slack, Teams oder die nächste Planung.
Dieser Bericht ist öffentlich. Dein Team kann ihn ohne Konto öffnen.
Gibt es Substanz, die ein Agent lesen kann?
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.
Server-rendered content detected (404 words, 2588 chars of visible text)
Low text-to-markup ratio (1.1%)
A very low text-to-markup ratio is a typical SPA-shell symptom. Inline more content directly into the HTML response.
Semantic landmarks present (main, article, section, header, footer, nav)
Single <h1> heading: "Where teams and agents Think together."
58/58 <img> tags have alt attributes
113/113 interactive elements have an accessible name
1/1 form controls are labelled
Method note: this reads markup, not a rendered accessibility tree
<title> length 45 chars: "The AI workspace that works for you. | Notion"
Meta description length 124 chars
Canonical URL: https://www.notion.com/
<html lang="en-us">
UTF-8 charset declared
Viewport meta present: "width=device-width, initial-scale=1"
Kann ein Agent die Seite überhaupt abrufen?
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 takes 2 redirects to answer
Every hop is a round trip the agent pays for, and some clients cap redirects well below a browser. Collapse the chain: point the first URL straight at the final one.
HEAD requests are supported
Content-Type: text/html with charset
Response compressed with gzip
No ETag or Last-Modified header — conditional requests unsupported
Send an ETag or Last-Modified header so crawlers can revalidate with If-None-Match / If-Modified-Since and receive a cheap 304 Not Modified instead of the full body.
Homepage size is reasonable (235.4 KB decompressed)
Roughly 647 tokens of content in 60,245 tokens of response (99% markup)
An agent pays to receive the markup and then discards it. Serving Markdown on Accept negotiation is the direct fix.
Homepage responded in 412ms
Meta-ExternalAgent is refused by Cloudflare while robots.txt permits it
Blocking it keeps your content out of Meta AI training and indexing. This edge verifies crawlers by IP range or signature, so the refusal may be correct anti-spoofing rather than a policy block. ax-audit sends this user agent from its own network without a Web Bot Auth signature, so an edge that verifies crawlers by IP range or signature will reject the probe while admitting the real crawler. Confirm against your WAF logs before changing any rule.
Amazonbot blocked at the server — consistent with its robots.txt Disallow
Bytespider is refused by Cloudflare while robots.txt permits it
Collects training data for ByteDance models. Compliance with robots.txt is disputed. This edge verifies crawlers by IP range or signature, so the refusal may be correct anti-spoofing rather than a policy block. ax-audit sends this user agent from its own network without a Web Bot Auth signature, so an edge that verifies crawlers by IP range or signature will reject the probe while admitting the real crawler. Confirm against your WAF logs before changing any rule.
2 crawler probe(s) could not be settled from outside
ax-audit sends this user agent from its own network without a Web Bot Auth signature, so an edge that verifies crawlers by IP range or signature will reject the probe while admitting the real crawler. Confirm against your WAF logs before changing any rule.
Site is served over HTTPS
HTTP requests redirect to HTTPS
HSTS max-age=31536000
HSTS includes subdomains
HSTS preload-eligible
Homepage is indexable
No directive restricts how AI assistants may use this page
Findet ein Agent, was Sie veröffentlichen?
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">.
OpenGraph required tags present (og:title, og:description, og:url, og:type)
Twitter Card required tags present (twitter:card, twitter:title, twitter:description)
/robots.txt exists
1/12 core AI crawlers configured
Add explicit User-agent entries for the missing crawlers with Allow: / for each one.
1 AI crawler(s) explicitly blocked
These crawlers have "Disallow: /" rules. If you want AI agents to access your site, change to "Allow: /" for each blocked crawler.
1 training crawler(s) blocked — recorded as a deliberate policy choice
Sitemap directive present
No Content-Signal directive found (optional)
Declare how crawlers may use your content after access with the Content Signals Policy, e.g.: Content-Signal: search=yes, ai-train=no. Known signals: search, ai-input, ai-train, plus the optional use=immediate|reference|full. Generate yours at contentsignals.org.
1/57 known AI crawlers have explicit rules
Add explicit User-agent entries for more AI crawlers to maximize discoverability.
Missing critical header: X-Content-Type-Options
Add the X-Content-Type-Options response header to your server configuration. This is a critical security header.
Only 3/7 security headers present
Add security headers like Strict-Transport-Security, X-Content-Type-Options, X-Frame-Options, and Referrer-Policy to your server response.
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.
/llms.txt exists
/llms.txt Content-Type OK (text/plain)
Missing H1 heading (first line should start with "# ")
Add an H1 heading as the first line of your llms.txt file, e.g.: # Your Site Name
Blockquote description present
7 section heading(s) found
48 link(s) found
/llms-full.txt not found (optional but recommended)
Create a /llms-full.txt with expanded content — full documentation, API details, and comprehensive site information for AI agents.
No rel="describedby" link to llms.txt
llms.txt v2 uses this relation so a page can name the file that covers it. Add <link rel="describedby" href="/llms.txt"> or the equivalent Link header, so an agent that landed on a deep page does not have to guess that an index exists.
3 duplicate link(s) in llms.txt
Each URL should appear once. Duplicates spend an agent’s budget re-reading what it already has.
15 sampled link(s) resolve
1 sampled link(s) redirect
Point llms.txt at the final URL. Each redirect is a round trip the agent pays for on every visit.
No per-page Markdown mirror found
llms.txt v2 documents appending .md to a URL for its Markdown version. The index tells an agent which pages exist; the mirrors are what make reading them cheap.
Consumer note: llms.txt is read by coding agents, not by search
Sitemap located: https://www.notion.com/sitemap.xml
Content-Type is XML (application/xml)
Sitemap-index references 169 child sitemap(s)
3/3 sample child sitemap(s) reachable
Sample yielded 274 URL(s) across 3 child(ren)
Newest <lastmod> is recent (17 day(s) ago)
Was kann ein Agent aufrufen?
/.well-known/agent-card.json not found
This site offers something an agent could call, but nothing tells an agent what. Publish an A2A Agent Card at /.well-known/agent-card.json. A minimal 1.0 card needs name, description, version, capabilities, supportedInterfaces, defaultInputModes, defaultOutputModes and skills. Spec: https://a2a-protocol.org/latest/specification/
API surface present but no machine-readable description found
The API exists; nothing describes it in a form an agent can read, so using it requires a human to read your documentation first. Serve an OpenAPI description at /openapi.json and advertise it with Link: </openapi.json>; rel="service-desc". For several APIs, publish an RFC 9727 catalog.
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.
No Agent Skills published
This site has documentation, so it has procedures worth teaching. A skill is a SKILL.md an agent installs and follows: setup steps, argument shapes, the mistakes to avoid. Publish one per task at /.well-known/agent-skills/{name}/SKILL.md and list them in /.well-known/agent-skills/index.json.
/.well-known/mcp.json exists
/.well-known/mcp.json is not a specified discovery path
Publish a server card at <your-mcp-endpoint>/server-card with Content-Type: application/mcp-server-card+json, and reference it from /.well-known/ai-catalog.json. Keep this file until your clients have migrated.
Valid JSON
Server name: "Notion"
Server description present
No tools array defined
Add a "tools" array. Each tool should have name, description and inputSchema.
No resources defined
Add a "resources" array listing the data your MCP server exposes.
No protocol version specified
Add a "protocolVersion" field declaring MCP spec compatibility. Current revision: 2026-07-28.
CORS enabled on MCP endpoint
No forms and no WebMCP code — nothing here for an agent to invoke as a tool
No commerce surface — agentic-commerce discovery does not apply to this site
Nothing on this site requires authorization — auth discovery does not apply
Welche Nutzungsrechte sind erklärt?
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.
/.well-known/security.txt exists
Required field "Contact" present
Required field "Expires" present
Expires date is in the future (2030-01-01)
2/5 optional fields present
Vollständige Berichte, priorisierte Korrekturen und Überwachung helfen deinem Team, Verbesserungen über das nächste Release hinaus zu erhalten.