# MistralAI-User

## Quick answer

MistralAI-User fetches a page only when someone using Mistral's Vibe assistant asks a question that requires visiting that specific site directly. Mistral says this bot is not used for automatic crawling of the web or for training its AI models.

## What it does

MistralAI-User activates when a person using Mistral's Vibe assistant asks a question that leads Vibe to visit a specific web page directly, rather than running as a standing background crawl across the web. Mistral is explicit that this bot is not used for crawling the web in any automatic fashion, and not used to crawl content for generative AI training, separating it clearly from MistralAI-Index, which does build a search index. Because each fetch corresponds to an actual user interaction with Vibe, the pattern of requests from MistralAI-User reflects what people are actually asking the assistant about, rather than a systematic sweep of a site. MistralAI-User respects robots.txt and Mistral publishes a dedicated IP range file for it, so a site can verify traffic and set rules specifically for this bot.

## Facts

- Operator: Mistral AI
- Robots token: MistralAI-User
- User agent string(s): Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; MistralAI-User/1.0; +https://docs.mistral.ai/robots)
- Purpose: user-triggered-fetch
- Respects robots.txt: yes
- IP ranges: https://mistral.ai/mistralai-user-ips.json

## Should you block it?

Blocking MistralAI-User mainly affects whether Vibe can successfully visit a site when a user specifically asks it to, for example by referencing that page or asking a question that leads the assistant there, rather than any broader crawling or training concern. A brand that wants to be useful when people ask Vibe about it, including through a direct question that would trigger a visit to its site, would generally want this fetcher allowed. Because Mistral states clearly that this bot is not used for automatic crawling or AI training, there is no training-data tradeoff here the way there is with crawlers like MistralAI-Index or other companies' training bots; the decision is narrowly about whether user-requested visits can succeed. A site with no connection to Mistral's user base has little reason to actively manage this bot either way.

## How to block it

robots.txt can include this rule to block MistralAI-User:

User-agent: MistralAI-User
Disallow: /

Mistral says this bot is used only for direct user-requested fetches in Vibe, not for automatic crawling or AI training.

## FAQ

### When does MistralAI-User visit a site?

Only when a person using Mistral's Vibe assistant asks a question that leads Vibe to visit that specific page.

### Does MistralAI-User train Mistral's AI models?

No. Mistral says it is not used to crawl content for generative AI training.

### Is MistralAI-User the same as MistralAI-Index?

No. MistralAI-Index builds a standing search index; MistralAI-User only fetches pages for specific user requests.

## Sources

- MistralAI-User is for user actions in Vibe and is not used for automatic crawling nor to train generative AI. Source: https://docs.mistral.ai/en/robots