Header Ads

Setup Browser AI Agent using Dockers

February 07, 2025 , ,
We now live in an age where AI tools have become essential components of our lives if we strive to be productive in our daily routines. Browser-Use is an open source AI agent that automates browser related tasks for example, if you on daily basis perform multiple browser related tasks such as searching products, videos, job hunting or even looking for booking tickets then this AI agent is for you as it automates all browser related tasks for you, so, you can sit back, relax and be more productive with your day.

Today, I shall be demonstrating setup process of this amazing AI agent Browser-Use using dockers on windows, so, you can right away start using it.




Prerequisites:

Before proceeding any further in this article, following are some of the many prerequisites for this article:
  1. Installation of Docker Desktop on Windows.
  2. Download of Browser-Use Code.
  3. Knowledge of Windows Command Prompt.
The running working executable batch file is provided to run this AI agent using docker on windows machine.

Download Now!

Let's begin now.

1) Make sure that you have install Docker Desktop on Windows properly without any errors.

2) After downloading and extracting the Browser-Use AI agent code from the GitHub repository to your designated path, use below command to change into the code root path i.e.

cd D:\BrowseUserAI\web-ui
D:

3) Now, use below command to create the ".env" environment file with the code directory if not already created i.e.

copy .env.example .env

4) Next step is to execute this AI agent with the help of below command i.e. 

docker compose up --build


5) Now, after docker completes building the tool's executable, copy the URL "http://localhost:7788" into chrome browser. Its important that you use chrome browser because by default the tool is setup for chrome browser, if you want to use another browser then you need to do the configuration through either the tool's UI or through ".env" file. You can look into the project GitHub documentation for it. You also need to provide your Learning Language Model (LLM) provider API Key such as openAI or Gemini or any other that is available within this tool.

You also need to open the VNC viewer in order to visualize the browser interactions of this AI agent within docker container because docker is a container, so, by default you do not have any browser installed within it that why you need to open the VNC viewer by copying the URL "http://localhost:6080/vnc.html" within the chrome browser. To connect to VNC server provide the default password "youvncpassword" mention in the tool's documentation. 


Without VNC viewer, you will not see the browser interactions but, if you click the result tab, you will able to see that the AI agent has been executed successfully and displayed correct result.


So here what you see within the VNC viewer, also as soon as you run the agent the browser automation begins within VNC viewer. i.e.













Conclusion

In this article, you will learn to setup this amazing AI agent Browser-Use using docker on windows. You also learn to configure LLM provider API keys as according to your choice and finally, you will learn to visualize the AI agent browser interactions within VNC viewer along with correct results within result tab.


Video Demo