browser-ai

What it does

Lets Jev choose bounded browser actions over Chrome CDP observations while a separate model writes form text.

How you can use it

Start by listing the exact steps a person takes on a website you want to automate. Write down simple tasks like clicking buttons, scrolling, or opening forms. A developer can use this plan to let Jev pick each next step from a limited list of numbered browser actions.

Your developer links the browser to this code and brings in a second tool only when typing is needed. Keep in mind that Jev does not write text. It only chooses set moves like clicking, scrolling, or waiting.