FREE Browser Use AI Agent🤖 OpenAI Operator Open Source Alternative Controls Your Browser!
In this video
- 00:00Introduction to Free Browser Use AI Agent: OpenAI Operator Alternative
- 00:27Overview of Browser Use AI Agent and Open Source Benefits
- 01:05Analyzing Browser Use Website and Web Agent Accuracy
- 01:36Cloud vs. Open Source Self-Hosted Version of Browser Use
- 02:13Getting Started: Cloning the Repo and Setting Up in VS Code
- 02:34Installing UV, Virtual Environment, and Dependencies
- 03:10Installing Playwright and Configuring API Keys
- 03:50Supported AI Models: OpenAI, DeepSeek, Anthropic & More
- 04:18Exploring Browser Use Documentation and Best Practices
- 04:56Using Cursor AI for Development and Custom Agents
- 05:48Running a Simple Task with Browser Use AI Agent
- 06:18Browser Use AI Finding Founders and Drafting Messages
- 06:56Generating a GIF of Browser Use AI Task Execution
- 07:36Testing Multiple Agents with Browser Use AI
- 07:59Finding YouTube Channel Info and Drafting Outreach Email
- 08:29Example Email for Video Editing Job Inquiry
- 09:09Sponsored Segment: Stride Agents AI Appointment Setters
- 09:46Creating Custom Agents and Testing Browser Use Features
- 10:19Using DeepSeek API for Reasoning in Browser Use AI
- 10:49Additional Use Cases: CAPTCHA, Job Applications, Social Posting
- 11:24Comparison with Other AI Agents: Open Operator, Agent Q & More
- 11:57Community, AI Academy & Business Coaching Opportunities
- 12:25How to Implement AI Agents in Business and Book a Call
- 12:48Final Thoughts: Keep Hustling & Accelerate Your Stride!
Links from the video
Transcript
2,582 wordsWhat's going on guys now on this channel we've covered many different AI agents that can control your browser and do specific tasks we've covered open interpreter we've covered agent e we've covered open operator we've covered agent Q we've covered open macro the list goes on and on and then recently most of you guys already know that open AI released open interpreter and this of course is an agent that can use its own browser to perform tasks for you and it's pretty
Impressive but unfortunately it is curent currently on open eyes Pro Plan only if you want to spend $200 a month then you could access this but if not you may want to use one of the open-source free versions which I'm going to show you in today's video in today's video we're going to be taking a look at browser use this is an AI agent that will be able to control your browser now the beauty about browser use is it isn't $200
A month and I'm going to show you how you can set it up in today's video Let's dive right into it all right guys so for taking a look at the browser use website first off I just want to point out that they have founder mode and VC mode here which is kind of funny that they're referencing that um but if we go ahead and look at the web agent accuracy we can see browser use right here um compared to some
Of the different ones Runner AG computer use web Voyager operator and it's doing pretty damn good and if if you go to the website you can see a more technical breakdown right here but we're going to dive into the actual tool so first things first they do have a cloud hosted version but it is 30 bucks a month so we're not going to be taking a look at this um we're going to be taking a look at their open- Source version
MIT license self-hosted full Library access all core features some of these features include Vision Plus HTML extraction multi-tab management element tracking custom action self- correcting any llm support so you can use this with deep seek open AI anthropic the list goes on and on now all links I cover in today's video will be linked down below so you can go over to their repo right here where we're going to get started and also check out their help section in their docs
Which I'll leave down below and we're going to go to the quick start section so the way we're going to get started is you're going to copy this repo URL and we're going to start by running get clone and then pasting in that repo URL depending on where we want to clone it then we're going to change directory into browser use and then we're going to open it up in vs code or cursor so I opened it up in cursor so
As you can see I have browser use opened up in our terminal now I'm going to go check out that doc section and we're going to do a few different things so first things first you're going to want to download UV if you don't already have it so simply go here click on this to actually download UV if you don't already have it on your system and you're going to run UV venv and then D- python so this is going to
Set up a virtual environment and then you're going to run either Source right here or VMV right here on windows so if you're on Mac or Linux you run this on Windows you run this to activate the virtual environment once you do that you're going to run UV pip install browser duse to install all the dependencies and then you're going to install playright by just simply running playright install so you can see here that I activated it then I ran browser
Use right here to install all the different dependencies and then right here I'm installing play right now in the dog section they go over actually how to create an agent right here so you can see different examples on how to create a basic agent how to set up your llm API key so you're going to want to make a copy of this env. example file and then you're going to want to paste in your open AI API key your anthropic API
Key and over here if we go to supported models you'll also see that there are other supportive models like azour Gemini deep seek right here and you simply just add in deep seekor API key right here Pas in your API key if you want to use deep seek like deep seek V3 deep seek R1 you can also use ol so you could use any single open source llm and then coming soon ones is Gro GitHub and fine tune models so they're
So they're improving this open source project which is really nice to see now if you're building something out for your custom use I would definitely like I said recommend checking out the dogs because it shows use things like different agent settings best practices browser settings connecting to your brows browser your output format system prompts sensitive data custom functions observability a bunch of different stuff so definitely check that out but just to show you some real world examples of how to actually
Use browser use we can actually leverage their repo right here because they have a bunch of different examples which will give you a really good um starting point to play around with this get a feel for it and then you can customize it for your specific use case also too I would recommend opening this up in something like Cur AI IDE of your choice so you could go over here to the chat right here you could chat like so and you
Could ask get specific questions or build composer and build out specific agents for you really good thing too is they also have their docs right here so you could leverage all this information right here such as local setup OBS availability customize right here so agent settings browser settings these are all U markdown files so you could easily plug this in uh reference this right here in cursor right here so we got browser agent agent settings all this good stuff and then
Easily get it to build you specific agents for your custom use case so I recommend referencing the docs here as well as potentially even different examples so how do you actually go ahead and run browser use well let's go to our first example so I'm going to go over to simple.pie right here so we're going to run Python and then simple.pie and we're going to click enter and you're going to see here the T askask is find the founders of browser
Use and draft them a short personalized message you can see here we're using opening eyes GPT 40 um we can see the different settings right here like the temperature we're loading the environment file we can see the agent and let's go ahead and look and now it's searching for founders of browser use we can see that it's getting a vision of all these different elements on the page here which is really cool that it shows you this visual right here so
You can see um this is the task so we can see all the different steps it took step one so starting to find the founders of browser use and draft a message next school search for the founders of browser use on Google action one out of one eval success found the information of the founders of browser use and now it's adding it to its memory right here so Magnus Muller and Gregor zunic next go draft the personalized message to these Founders
And then boom you can see our message right here so congratulations on your success with browser use dear Magnus and Gregor I hope this message finds you blah blah blah blah blah best regards your name so task completed successfully created GIF at agent. GI so this is another really cool thing it does if we go over here to agent. his.gf we can see a gif of our specific task so find the founders the browser use so step one search for the
Founders draft a personalized message in step two and then boom so pretty straightforward and simple now let's go ahead and do another one so this time I'm going to use this one which is python multi multiple agents same browser. py and we're going to run this okay so we can see logging setup okay so we can see for this one it's actually asking us for a task so I'm going to go ahead and enter a task here okay so here's our
Task go to YouTube search Josh pook and find out how many subscribers he has then craft a personalized Outreach email to ask him to hire me to be his video editor and reference specific stuff from the titles of some of his most recent videos let's go ahead and send this okay so now we can see Step One open tab YouTube so if we go here we can see that it opened up YouTube all right now it's clicking on the element and
It typed in Josh pook right here okay now it clicked on search so it's actually loading up Josh pook now it's analyzing the page again okay and boom we got our task completed here so let's go ahead and take a look so we can see that it grabs some of these titles right here this one mentioning deep seek R1 n8n this one is about site Brew right here so subject elevate your video content with professional video editing hi Josh I hope
This message finds you well I'm reaching out to express my admiration for your exceptional work especially your recent videos like build no code AI agents with deeps R1 in N8 an are you tired of pouring thousands of dollars into appointment Setters only to watch leads slip away imagine having a team of elite sales agents booking qualified appointments for you around the clock no more wasted time on training no more frustration with performance and no more draining your budget on inconsistent and
Expensive call centers introducing stride agents AI powered appointment centers that work 24/7 never get tired and book appointments while you sleep trained on thousands of successful conversations our AI agents El performed human teams at just 110th of the cost join the ranks of businesses that doubled their appointments and booking rates in just a matter of weeks don't get left behind in the AI Revolution visit stride agents.com now and transform your entire sales process with Cutting Edge AI technology it's time to
Accelerate your stride with AI agents and new site Brew AI coding agent vzer your insightful content from your community of 16.5k subscribers is truly inspiring I specialize in video editing and believe I could enhance your visual storytelling of your content my personal experience aligns with your Niche and I'm eager to help your channel grow further I would love the opportunity to discuss how I can contribute to your future Projects please let me know a convenient time for us to connect looking
Forward to an opportunity for us to work together warm regards lank all right so keep in mind this was a very basic prompt like you could personalize this like and make it very specific and feed this information like if I was using this in a real world production scenario I would give it a lot more context and build a more custom agent but I hope that this kind of Sparks some creativity for you guys as you can see there's a bunch
Of different options here like validate output sensitive data results processing we got a planner agent right here so if we check the planner out you can see that it's referencing deep seek right here so deep seek API so what you would want to do is simply just add this deep seek API in your. EN uh environment file right here and then we just simply change our task right here to a specific task like search Josh poock and tell me what he
Does and then run python planner. piy now you can see it's either going to use deep seek API key right here for uh deep seek model or it's going to use 03 mini so we're using a reasoning model right here and you can see the actual reasoning right here so the current browser state is minimal with an empty page next steps determine the intended web page or content to to load and I'm not going to go through this whole entire one
But you can see here the actual reasoning right here so you can see the thinking process and everything like that so the current state provides no interactive content only an empty blank page the next step is to do so and so so you go ahead and run these test this out you can see download file right here you can see custom functions browser you can see different use cases like capture check appointments find and apply to jobs online coding agent right
Here post on Twitter scrolling page shopping the list goes on and on so that's pretty much it for this video guys on browser use let me know down below if you've been trying different AI agents that are browsing the web like open ai's open operator or agent Q agent e or browser use let me know which one of these is your favorite and let me know what you think about browser use guys in the comments down below if you're new to
The channel we upload videos all the time on ai ai agents AI coding Marketing sales business growth so if you like that type of content you got some value here make sure to like the video comment down below and subscribe to stay up to date with the uploads also too guys if you haven't already joined our free Facebook group and Discord Channel strid community.com I'll leave a link down below for that and then also too guys you're definitely going to want
To check out our stride AI Academy for behind the scenes resources templates tools trainings Etc that I don't post here on YouTube and then also too guys if you run a business and you need help implementing AI agents into your business or if you want coaching and training on how to actually sell these AI agents to other business owners and make 2 to 10K plus then book a call down below at executiv st.com slapp and we can see if it's a
Potential fit to work together or not other than that guys I will see you in the next video keep hustling keep grinding and of course guys accelerate your stride take care