AEO grade · mistral.ai

Will AI engines citemistral.ai?

Quick grade · 62/100

mistral.ai has serious AEO gaps. AI engines may not be reading or trusting your pages the way you think. The fixes are mostly cheap — schema, llms.txt, and crawl-time signals.

X LinkedIn

What we checked

HTTPS10 pts
TLS responds correctly. AI crawlers won't be blocked at the network layer.
robots.txt + AI bots15 pts
robots.txt exists but doesn't mention any AI engine bots. Default behavior is allow, so you're crawlable — but you've left a high-leverage signal off the table.
llms.txt10 pts
llms.txt found. You're telling AI engines exactly which content to ingest — a 2026 frontier signal.
sitemap.xml10 pts
No /sitemap.xml at the standard location. AI engines fall back to link-following, which under-discovers deep pages.
Open Graph image10 pts
og:image is set. Social and AI-engine link previews will render with a branded card.
JSON-LD schema20 pts
No JSON-LD on the homepage. Without schema, AI engines fall back to plain text parsing — far less reliable for facts, prices, ratings, and entity matching.
<title> length5 pts
Title length (56 chars) sits in the AI-Overview-friendly 50–60 char band.
meta description5 pts
Meta description (149 chars) sits in the 80–160 sweet spot.

Quick grade checks 8 signals at the homepage level. The full VectorCite audit scores 47 signals per page and runs against real buyer queries you specify.

Go deeper.

The full 47-signal audit scores your page against the buyer queries your customers actually type — and outputs the exact fixes to close the gap.

Run full audit · free