TelegramBot: Telegram Link Preview Fetcher User Agent

TelegramBot is the name that Telegram’s link preview fetcher sends in its requests. It is not an AI crawler. Telegram fetches a shared URL to show a link preview, and sometimes to build an Instant View page. Telegram documents Instant View, but it does not publish the user-agent string or a robots.txt policy.

User-agent string

A user agent is the client name in the User-Agent HTTP header. Telegram does not publish its user-agent string. Public bot lists, such as crawler-user-agents, record this string for Telegram link previews:

TelegramBot (like TwitterBot)

If you match it in logs or server rules, match TelegramBot and ignore case.

Link previews and Instant View

A link preview is the card that a messaging app shows for a shared link. Telegram’s Instant View page explains how the two features connect. When Telegram makes a link preview, its Instant View bot looks for an Instant View template for that website. A template is a set of rules for one domain or URL.

If a template exists, the bot fetches the page and builds a simple version of it inside Telegram. The Instant View docs say that the bot processes only pages with the MIME type text/html. Anyone can make a template in the Instant View Editor. Telegram’s team approves a template before all users get it.

Telegram says that site owners do not need to change their sites for Instant View to work. Telegram does not document which meta tags its link previews read. Other preview services, such as Snapchat, document that they read Open Graph meta tags.

Block TelegramBot

Telegram does not say whether its fetcher reads robots.txt. If you want to stop the requests, block the user agent on your web server. This nginx example returns HTTP 403 for any user agent that contains TelegramBot:

if ($http_user_agent ~* "TelegramBot") {
    return 403;
}

Put the example in the server block of your site configuration, then reload nginx. Without access to your HTML, Telegram cannot build a preview or an Instant View page from your page.

Verify requests

Telegram publishes no IP list and no reverse DNS method for this fetcher. Anyone can copy a user-agent string. A match in your logs does not prove that the request came from Telegram.