Sign in
Start free

AI agent directory / MistralAI-Index

MistralAI-Index

Crawler from Mistral

Crawls pages so they can appear in Mistral search, which Vibe uses to answer questions.

MistralAI-Index is Mistral's automated web crawler. Mistral says it crawls for indexing only: the pages it fetches feed Mistral search, which helps answer user questions in Vibe.

Mistral states that content crawled by MistralAI-Index is not used for generative AI training of any kind. It is separate from MistralAI-User, which fetches a page only when a user asks, and from MistralAI-Training, which collects training data.

Run byMistral
TypeCrawler
PurposeSearch
Acts forIts operator
robots.txt tokenMistralAI-Index
IP rangesPublished by Mistral
DocumentationMistral's documentation

MistralAI-Index user-agent string

Full user-agent string (source):

Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; MistralAI-Index/1.0; +https://docs.mistral.ai/robots)

Match on MistralAI-Index; the rest of the string changes between versions.

Verifying it's really MistralAI-Index

Mistral publishes the IP addresses MistralAI-Index uses at mistral.ai/mistralai-index-ips.json. Check that a request with the MistralAI-Index user agent comes from an address on that list.

How Airglow identifies it

  • DeclaredIt names itself in the User-Agent: MistralAI-Index.

Every visit in Airglow says which of these applied, so you can decide how much to trust it.

Seeing it on your site

It usually reads pages without running JavaScript, so Airglow counts every visit through a server connection (Cloudflare). When it does run JavaScript, the script sees and replays it too.

What the script and a server connection each see

Allowing or blocking MistralAI-Index

MistralAI-Index reads the MistralAI-Index group in your robots.txt. To keep it out of the whole site:

User-agent: MistralAI-Index
Disallow: /

Airglow shows, for every agent that visits, whether your robots.txt lets it in, and helps you write rules by purpose (training, search, acting for a person).

Mistral's three robots.txt tokens

Mistral documents three separate user agents, each with its own robots.txt token:

  • MistralAI-Index crawls automatically to build the Mistral search index. Not used for training.
  • MistralAI-User fetches a page when a Vibe user asks a question. Not used for automatic crawling or training.
  • MistralAI-Training crawls to build datasets for training Mistral's generative models. Not used for search or live answers.

Because the tokens are separate, you can block training while still letting your pages appear in Mistral search, or the reverse.

Questions

Does MistralAI-Index collect data for training Mistral models?

No, according to Mistral. Its documentation says content crawled by MistralAI-Index is not used for generative AI training of any kind. Training crawls use a separate agent, MistralAI-Training.

What happens if I block MistralAI-Index?

Mistral's crawler will stop indexing your pages for Mistral search, which is what Vibe draws on to answer questions. Blocking it does not block MistralAI-User or MistralAI-Training, which have their own tokens.

How is MistralAI-Index different from MistralAI-User?

MistralAI-Index crawls on its own schedule to build a search index. MistralAI-User visits a page only when a person asks Vibe something that needs it, and links the source in the answer.

Does Mistral publish IP addresses for MistralAI-Index?

Yes. The list is at https://mistral.ai/mistralai-index-ips.json, a JSON file with a creationTime field and a prefixes array of ipv4Prefix entries.

Other agents from Mistral

Sources

Last reviewed November 10, 2026.

See when MistralAI-Index visits your site

Airglow names every AI agent on your website, replays what it did and links the visitors it sends you. Free for one website.

Start freeAll agents