PanguBot: Huawei AI Crawler, What Is Known and Blocking

PanguBot is a user agent that third-party bot lists connect to Huawei and to the training of its PanGu AI models. Huawei does not publish documentation for a crawler with this name. If you do not want it on your site, add a robots.txt group and also block the token on your server.

What is known

Huawei documents one web crawler, PetalBot, for Petal Search. A search in October 2026 found no Huawei page that describes a crawler named PanguBot. There is no published user-agent string, IP list or robots.txt statement from Huawei.

Third-party bot lists fill the gap. For example, Known Agents says that PanguBot downloads training data for PanGu, Huawei’s multimodal large language model. These lists do not agree on the full user-agent string. Known Agents lists this one:

Mozilla/5.0 (compatible;PanguBot;[email protected])

Other lists show different strings. All of them contain the token PanguBot.

Do not confuse it with Pangu Bot on Huawei Cloud

Huawei Cloud sells a chatbot service named Pangu Bot. Its service overview describes a dialogue platform that companies use to build question-and-answer chatbots. That product is not a web crawler.

Block PanguBot in robots.txt

robots.txt is a text file at the root of your site with rules for crawlers. Crawlers match groups by name without regard to letter case, as RFC 9309 requires. If you want to block PanguBot, add this group:

User-agent: PanguBot
Disallow: /

Huawei does not say that PanguBot obeys robots.txt. RFC 9309 also states that robots.txt rules “are not a form of access authorization”. If you want to make sure that the requests stop, also block them on your server.

Block PanguBot on the server

This nginx example returns HTTP 403 for any user agent that contains the token:

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

Put the example in the server block of your site configuration, then reload nginx.

Verify requests

Huawei publishes no IP list and no reverse DNS method for PanguBot. You cannot prove that a request with this token comes from Huawei. A user-agent block stops all requests with the token, real or not.

What a block changes

A PanguBot group in robots.txt does not apply to PetalBot, because each crawler uses its own token. A block of PanguBot alone does not remove your site from Petal Search.