Platform playbook

AI-agent readiness for Framer, Wix and Squarespace sites

Visual site builders solve the design problem and then own the parts a crawler cares about. Text lives inside layout components, robots.txt is generated by the platform, and arbitrary files at the root of the domain are usually not offered at all. That does not make the score unfixable — it moves the work into the settings panel and the page structure rather than into a code editor, and it makes it worth knowing which points are genuinely out of reach before you chase them.

What our own scans say about Framer

Not an estimate. These numbers come from public AIScan reports and change as we scan more.

Framer sites scanned
6

Distinct public domains, counted once each at their most recent scan.

Average score

Only 6 public Framer sites so far. That is too small a sample to publish an honest average, so we are not printing one.

What Framer sites fail most often

Computed from public AIScan reports, refreshed as we scan. Private reports are never included.

How Framer sites fail, and what fixes it

Every heading is a styled text layer

Canvas builders happily let you make a caption look like a heading and a heading look like body text. The served HTML then has three h1s or none, which is exactly what check C3 and the heading hierarchy in E3 penalise.

C3 · Structured HTML (title, meta, JSON-LD, single H1)E3 · Heading hierarchy & server-rendered text

No way to place a file at the root

llms.txt and .well-known files need to sit at the domain root. Where the builder cannot serve them, the honest options are a redirect from a page you can control, a proxy in front, or accepting the deduction and winning the points that are reachable.

C2 · /llms.txtP1 · API Catalog (RFC 9727)

Platform robots.txt with no AI stance

You typically get a toggle for indexing and nothing about AI crawlers. Where the builder exposes custom robots rules, that panel is the only place AI bot rules will survive a republish.

D1 · robots.txt present & saneB2 · Explicit AI bot rulesB1 · Content Signals in robots.txt

Content that only appears on scroll

Reveal-on-scroll and lightbox patterns are fine for decoration and fatal for facts. Anything a visitor would quote — price, hours, spec — has to be in the initial HTML.

E3 · Heading hierarchy & server-rendered text

Step-by-step fixes: /docs/platforms · /docs/checks/content · /docs/checks/bot-access · /llms-txt-generator

Businesses running on Framer

Questions Framer owners ask

Can I reach 100 on a hosted site builder?
Rarely, and we would rather say so. Root-level files and headers are the usual ceiling. Grade A is normally reachable; chasing the last few bonus points is not a good use of your afternoon.
The builder generates robots.txt. Can I still add AI rules?
Only if it exposes a custom rules field — several now do. If it does not, the deduction on B2 is a platform limit and the report says so rather than blaming you.
How do I fix headings without redesigning?
Change the semantic tag, not the styling. Every builder separates the two; it is usually a dropdown on the text layer.
Is a CMS collection page better than a static page?
Yes, generally — collections produce consistent markup and a sitemap entry per item, which is more than a hand-placed page usually gets.
Should I move off the builder to score better?
Not for this. Scan first, fix what the panel allows, and only consider moving if the missing points actually matter to how you are found.

Scan your Framer site and see where you actually stand

Free, no account, about twenty seconds. The report names every failing check by ID, shows the evidence we found, and gives the fix for your platform — plus a hand-off prompt you can paste straight into Claude Code or Cursor.

Run a free scan npx aiscan-cli yoursite.com

Prefer to read first? Browse every check we run or the guide library.