The Grammage robot

What the robot does

When someone runs a free test, an audit or a measure in their CI/CD, Grammage opens the page in a real Chrome browser and loads it the way a visitor would, to measure its weight, its requests and the carbon footprint that follows. That browser is what your site sees.

It only visits the pages it is asked to measure, one at a time, and it never fills in a form. If your site protects its pages against robots, it is normal for it to be stopped, and it is up to you to decide whether it may pass.

Recognising it

Its user agent is that of a real Chrome, mobile or desktop depending on the measure, with the Grammage token added at the end, followed by its version and a link to this page, for example:

Mozilla/5.0 (Linux; Android 10; K) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/<version> Mobile Safari/537.36 Grammage/<version> (+https://getgrammage.com/robot)

Allowing it

The steps depend on the protection that filters your site, and in all three cases they take a few minutes from its console.

DataDome. In the console, add an allow rule on the user agent containing Grammage.

Cloudflare. In the security rules of the zone, create a custom rule with the “Skip” action, on the user agent containing Grammage, and tick the bot protections it should let through.

Akamai. In Bot Manager, declare a custom bot recognised by the Grammage token of the user agent, then give it the “Allow” action.

Contact us

For any question about this robot, or if you would rather it stopped measuring your site, write to us at [email protected].