Mistral · Search index
MistralAI-Index
Scheduled sweeping is the job: Mistral calls it "automated crawling of the web for indexing purposes only" and says it "indexes content for Mistral search, which helps answer user questions in Vibe." Same destination as `MistralAI-User`, opposite timing: this one builds the corpus before anybody asks, where the user token fetches the single page a live question needs. The training carve-out is phrased more strongly than most vendors manage: content crawled here is "not used for generative AI training of any kind." Its range file is the freshest in this group, timestamped April 2026, but it declares only two hosts. Two prefixes is a thin fleet for a general web indexer, so an IP check will reject legitimate requests the moment Mistral scales without republishing.
- Operated by
- Mistral
- Purpose
- Search index
- robots.txt token
- MistralAI-Index
- Verification
- IP verified
Crawls continuously to build and refresh an AI search index.
How to verify MistralAI-Index
Mistral publishes the IP ranges MistralAI-Index crawls from, and we fetch that list on a schedule. A request claiming to be MistralAI-Index can therefore be checked against 2 published ranges: one that does not match is not this crawler.
Published ranges last confirmed by us on .
User agent
Mistral publishes this user agent for MistralAI-Index. Match on the MistralAI-Index product token rather than the whole string: vendors revise the surrounding version and URL fragments without notice.
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; MistralAI-Index/1.0; +https://docs.mistral.ai/robots)robots.txt for MistralAI-Index
Controllable via the `MistralAI-Index` robots.txt token, under Mistral's page-level statement that it "uses specific `robots.txt` tags" so webmasters can "manage how their sites and content interact with AI." No exemption or override is claimed for it. Mistral publishes no explicit compliance sentence for this token either, so the accurate description is that the directive is the documented control, not that compliance is guaranteed in writing.
Block
User-agent: MistralAI-Index
Disallow: /Allow
User-agent: MistralAI-Index
Allow: /robots.txt is a request, not an enforcement mechanism. It is honoured by convention, and a crawler that ignores it is stopped at your edge, not in a text file.
What blocking MistralAI-Index costs you
Disallowing it empties you out of the corpus Mistral's search reaches into, so Vibe has nothing pre-indexed to offer when a question touches your subject: you stop being a candidate answer rather than losing a formatting nicety. The reach is wider than the user token's: that one only matters once a question already names your page, while this one decides whether Mistral can surface you unprompted at all. No training protection comes with it, since Mistral excludes this crawl from generative training "of any kind"; `MistralAI-Training` is the token that governs that, on its own separate directive. Blocking Index while leaving Training allowed inverts what most operators actually intend.