MistralAI-User
Mistral AI · User fetch · robots.txt: Yes
Visits a web page when a user asks Mistral's Vibe assistant a question, to help answer it and link the source.
Block MistralAI-User
User-agent: MistralAI-User
Disallow: /
Blocking MistralAI-User tells Mistral that Vibe should not fetch your pages when a user asks. Mistral says this agent does not crawl the web automatically and does not collect content for generative AI training.
Allow MistralAI-User
User-agent: MistralAI-User
Allow: /
An allow group only changes anything if a broader rule would block MistralAI-User. Under the robots.txt standard (RFC 9309, section 2.2.1) a crawler follows the group that names it and uses the User-agent: * group only when no group does — so this group lets it in even if your * group says Disallow: /. Token matching is case-insensitive.
Put these lines in /robots.txt at the root of each host (each subdomain has its own file). robots.txt is a request to well-behaved crawlers, not access control.
Already have a robots.txt? Paste it into the robots.txt checker to see whether it blocks MistralAI-User on a given path, and which line decides.
Block or allow it at the server
robots.txt asks; a server rule refuses. Every rule here matches MistralAI-User, spelled as in the User-Agent Mistral AI documents (“compatible; MistralAI-User/1.0;”), and matches it case-sensitively. The server answers 403.
What a User-Agent rule cannot do: any client can send any User-Agent. Access control by User-Agent is an unreliable technique, since the User-Agent header can be set to anything at all, at the whim of the end user.
(Apache HTTP Server 2.4 documentation). A fake MistralAI-User is refused by the block rule along with the real one, and a crawler that hides its name passes. The “only the real one” rule is the check: it lets the name in only from Mistral AI's published addresses.
Addresses in the “only the real one” rules: Mistral AI's list mistral.ai/mistralai-user-ips.json, 4 ranges, fetched 8 Oct 2026, list dated 19 Feb 2025. Operators change their lists; compare the date with the live list before you rely on it.
The IP rules compare the address your server sees. Behind a CDN or load balancer that is the proxy's address, so every claim would be refused: restore the client address first, or put the rule at the CDN.
Syntax from each server's own documentation, fetched 9 Oct 2026: nginx (rewrite module, map module, geo module); Apache (Access Control, <If>, expressions, mod_authz_core, mod_authz_host); Caddy (request matchers, respond); Cloudflare (custom rule example, operators, values and lists, http.user_agent). Every line form above is listed with its quote in server-rules-sources.json.
Who blocks it
Of the 1,000 most-visited domains on Tranco list PY35J (generated 2026-10-08), we read 508 robots.txt files on 2026-10-09 and ran each through the same RFC 9309 matcher as the checker, for the path /. 33 block MistralAI-User from the whole site by name, 34 block it through their User-agent: * group, and 5 name it with partial rules (some paths closed, the home page open). 2 more name it only to let it in.
33
block it by name 6.5%
34
block it through * 6.7%
5
name it, partial rules 1.0%
Percentages are of the 508 files we could read, not of all 1,000 sites.
Highest-ranked sites that block MistralAI-User by name
- amazon.com rank 27
- tiktok.com rank 58
- msn.com rank 72
- amazonvideo.com rank 111
- cnn.com rank 205
- sciencedirect.com rank 223
- amazon.co.uk rank 336
- amazon.de rank 400
- amazon.co.jp rank 491
- uol.com.br rank 556
Each links to the site's live robots.txt, which may have changed since 2026-10-09.
robots.txt is only what a site asks for. A site can also turn a crawler away at its firewall or CDN, or let an operator in under a licensing deal, and neither shows in its robots.txt. A block through * stops every crawler the file does not name, so it says nothing about this one in particular. The files we could not read are not counted: 222 of the 1,000 domains are infrastructure names with no website at the bare domain (CDN and DNS zones, for example), 59 answered our request with 401 or 403, 93 failed (timeout, TLS or server error), 59 have no robots.txt (404 and similar) and 59 answered with an HTML page instead of a robots.txt file. Every agent's count, side by side.
Facts
- Operator
- Mistral AI
- Purpose
- User-requested fetch
- Obeys robots.txt
- Yes. Mistral says MistralAI-User governs which sites these user-requested fetches can be made to.
- In your server logs
- The user-agent string contains "compatible; MistralAI-User/1.0;".
- How to verify it
- Match the IP against Mistral’s published MistralAI-User IP list.
Published list: mistral.ai/mistralai-user-ips.json — 4 IPv4 ranges when we fetched it, list dated 19 Feb 2025. Always match against the live list; operators update them. To count real and fake MistralAI-User requests in your own server log against this list, drop the log into the log reader.
What Mistral AI says
Quoted word for word from the operator's documentation.
“MistralAI-User is for user actions in Vibe. When users ask Vibe a question, it may visit a web page to help answer and include a link to the source in its response.”
“MistralAI-User governs which sites these user requests can be made to. It is not used for crawling the web in any automatic fashion, nor to crawl content for generative AI training.”
“Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; MistralAI-User/1.0;”
“All published IP addresses: MistralAI-User IP addresses”
Sources
- Mistral Docs: Mistral crawlers — fetched and checked 2026-09-24
- Mistral: MistralAI-User IP addresses (mistralai-user-ips.json) — fetched and checked 2026-09-24