In short. YandexUserproxy is documented for user-triggered service requests and online page translation.
Definition: YandexUserproxy is a Yandex robot that proxies user actions on Yandex services. Yandex documents it as sending requests after button clicks and downloading pages for online translation, so a request with this user-agent can follow a user action without being a regular indexing visit.
Key facts
- Yandex lists this full user-agent:
Mozilla/5.0 (compatible; YandexUserproxy; robot; +http://yandex.com/bots). Yandex's robot reference gives its purpose as proxying user actions and downloading pages for online translation. - The same reference marks
YandexUserproxyas not taking the general rules specified inrobots.txtinto account. - Yandex says some special-purpose robots can ignore restrictions written for
User-agent: *, and recommends directives for the exact robot when access must be restricted. - Yandex Translate's site mode asks a user to paste a website link, then displays the website translation after the site action is selected. Its website translation documentation describes this flow.
Why does it appear after a button click?
Yandex's robot reference describes YandexUserproxy as a request proxy for Yandex services. Its purpose field specifically says that it sends requests in response to button clicks. A server log entry with this user-agent after a click therefore fits a user-triggered service request. The same purpose field also names page downloads for online translation, which accounts for visits that begin when a translation control is used.
The user-agent is distinct from the main search crawler in Yandex's table. The table identifies YandexBot/3.0 as the main indexing robot, while the YandexUserproxy row describes service actions and shows No in the column for general robots.txt rules. The request's purpose, rather than its position in a crawl sequence, is the relevant signal in this case.
What does translation have to do with the request?
Yandex Translate's website translation documentation documents two website translation modes. In text mode, a user pastes a website link into the translation field. In site mode, the user selects Site and pastes the link. Yandex says the service automatically detects the source or target language when the language is not selected manually.
The resulting page fetch is consistent with the robot reference's description of downloading pages for online translation. It is a request made for a service response. The reference does not describe a separate URL pattern or a fixed delay between the click and the request, so logs should be read from the user-agent and surrounding request context rather than from timing alone.
How should access rules be checked?
Yandex's robot logs documentation says that some robots that download documents for purposes other than indexing may ignore restrictive rules written for arbitrary robots. Its user-agent guidance says to specify the exact user-agent for robots that may ignore the broader User-agent: Yandex group. A matching rule can be written in this form:
User-agent: YandexUserproxy
Disallow: /
This rule targets the token shown in the documented full user-agent. Apply it only when the site's service requirements allow the request to be restricted, because it can affect Yandex service actions and online translation requests. Yandex's example for special-purpose robots uses the same exact-user-agent structure.
To verify that a log entry belongs to Yandex, start with the IP address recorded by the server. Yandex instructs webmasters to perform a reverse DNS lookup, check that the host name ends in yandex.ru, yandex.net, or yandex.com, and then perform a forward lookup whose address matches the original IP. Yandex also says that its bot addresses change frequently and that the full list is not disclosed.
What should multilingual sites expose?
Yandex's localized-pages documentation says to specify alternative language or regional versions for the indexing bot. It recommends localized page markup because Yandex no longer supports using a Sitemap file for language versions. Each alternative should be a complete and accurate translation of the original page.
For a page whose language is selected from user settings, an Accept-Language header, or an IP address, Yandex recommends an x-default alternative. Its example uses this form:
<link rel="alternate" hreflang="x-default" href="http://example.com/" />
Related terms
FAQ
Is YandexUserproxy the main Yandex indexing crawler?
No. Yandex's robot table names YandexBot/3.0 as the main indexing robot and gives YandexUserproxy a service-request and translation purpose.
Does a YandexUserproxy request prove that a page was indexed?
No. Yandex documents this robot for purposes other than indexing. The page also states that a document inaccessible to the main Yandex robot will not be indexed when downloaded by such a robot.
Should a site block the user-agent?
That depends on whether the site needs Yandex service actions or online translation. If access must be restricted, Yandex documents exact-user-agent directives for special-purpose robots.
Sources
- Yandex Webmaster, "How to check that a robot belongs to Yandex", undated
- Yandex Webmaster, "The User-agent directive", undated
- Yandex Translate, "Translating websites", undated
- Yandex Webmaster, "Indexing localized pages", undated
Check how a page reads to answer engines. Open the GEO score analyzer