NEW Hermes AI Browser Agent: Automate ANYTHING?

Hermes agent just got a massive free upgrade and it turns your AI agent into something that can actually browse the web and work for you. I'm talking about booking things, finding leads, tracking data, filling out forms, all done by your AI agent Hermes automatically whilst you're not even at your computer. This works because Hermes now has access to a free library of 250 browser skills. Skills built for the biggest websites on the planet and your agent can now use any of them the moment you ask.

And in this video I'll show you the exact setup, the exact commands. I'll show you the one thing you have to do first because most people set this up wrong and then wonder why their agents keep getting stuck. So stay with me because by the end of this you're going to have a browser agent workforce running for you 24-7. Let's go.

There's a brand new update from Hermes with these new browser agents and essentially Hermes agent now has access to hundreds of browser skills through browser base right. So that means your AI agents can use the browser better than ever before using Hermes agent and this powerful stack that I'm going to guide you through today. This is a super powerful system because basically what you have here is a whole catalog of browser agents that you can use to automate whatever you want and it's really powerful and easy to install and you can also browse through everything like you see right here and we're going to guide you through exactly how it works, how to set it up, how we use it etc and how you can really make the most out of these new browser agents inside Hermes. Now you could say for example okay plan a road trip to Utah with all these stops in between and organize where we're going to go and then book and get that sorted out right here and you can actually do that with a browser command right.

So it's basically a way to empower your AI agents to browse the web, they can do bookings for you, they can automate a lot of tasks like having a personal assistant that you can speak to and it works 24-7, it can automate tasks for you, saves a lot of time. So how does this work, how do we do it etc. So the way that you can set this up is you can actually go inside Hermes agent. If you don't already have it installed you can just go to the github and get it set up and then from here what you want to do is use this command right npm install gbrowse right.

Once you've done that basically browse will now be installed and you can use it to install skills from the browse.sh catalog right. So you can browse the list to see available skills and you can browse install to get set up right which is pretty amazing and you can also pull in and install skills whenever you want. So what these skills you can set up with Hermes agent, what sort of stuff can it do, that's a good question. So what you can actually do is browse all this stuff here and you can see for example you can like search and you can search like ebay for you and stuff like that.

You can you can search the web, it can find products for you, you can basically automate your shopping for you, you can have a look across the browser, really powerful stuff right and any of these skills you can actually pull in directly into your AI agent. So for example if we wanted to set this one up we can just grab that and then go into Hermes agent and go from there right. It's pretty powerful and so I'm going to talk about a simple system that lets one person command an army of AI browser agents from a single command line right with no code, no technical background needed, just one screen, one method and agents that can go off and do things for you. So imagine for example you had a tiny little AI agent that could browse the web and do work for you.

It could for example check flies, it could track packages, it could get leads for you, it could fill forms, it could post content, could find deals for you, could pull data all by itself right. That tiny AI agent is called a browser agent and so now imagine like you had 50 of them all different doing different jobs at the same time. It sounds great but it would also be chaos right. You would lose track of who was doing what, you would forget which agent was already running, you would have 50 browser tabs open and no idea what was going on right and that's where agent operating system comes in as well.

So agent operating system is like a control tower for all your AI agents with one screen, one command line and every agent reporting back to you in one place which is what we've got over here right. This is agent operating system. This is something that I've built out. You can get the full training inside the AI Profit Boarding if you want the full system and files and prompts from me or what you can actually do is set up yourself but it might be quite time consuming.

I would recommend just going with ready to go setup. So let's break this down in terms of what it means and how it works. So you got Hermes which is a chat window that controls the AI agent. Then you got Browser.sh which is a free library of 250 pre-built browser skills right which works across like all the biggest websites in the world and then you have agent operating system which is the system that ties them all together so you can manage everything from one home base right and the way that I see this is like most people use AI like a calculator but a small group have learned to use AI like a team, like a workforce right and the conductor stack is how you cross that gap.

Now you might be wondering what is a conductor stack. I'm going to guide you through that in a second. Let's see what questions we've got here. So Rob, shout out to you, you legend, says I love all of your training videos.

It's my first time live. Fantastic to hear about. Thank you for joining. If anyone's watching this feel free to ask questions as you go along.

That's what I'm here for. Let me fill in the gaps for you if you have any questions as well. So you know always here to help you. So what is the conductor stack?

This is a simple model behind everything you're about to learn right. You want to think of yourself as the conductor of an orchestrator of an orchestra right. So you don't play the violin, you're not going to be playing the drums right. You just raise your arms and the musicians play right and this is what you're about to become but with AI agents instead of musicians.

So the conductor stack has five layers and each layer feeds into the next. If you skip a layer the music falls apart. So you want to take this pretty seriously. So you've got for example the conductor, you've got Hermes, you have the orchestra which is browser skills, you have the score which is agent OS and you have the encore.

All right so let me walk you through each part. So layer one, the conductor. You set the vision and this is you. You set the vision and the rules right.

You decide what gets done. You never touch a violin, you just point and approve. That's what we're doing here. So for example if we go into Hermes over here right you can see here I've said hey you have browser agents right.

It says yes I have browser tools you want me to do anything here and so what we can actually do is as an example of this we can say okay give me a list of what you can do right and we can actually get Hermes to give us a list and then we're just going to mention for example the browse setup and we can speak to Hermes inside our chat. By the way like how much nicer does it look using this versus this. For 99% of people they just don't know how to use a terminal and it's quite intimidating and then number two you don't get any of your chat history and everything else right and so that's why you want to build an agent operating system. I highly recommend it right.

So you can see here that actually gives you all this stuff that you can do. So you can browse pages, you can take snapshots, you can for example you can click around, you can navigate, you can also use your mouse, you can browse sessions, you can actually manage your sessions with these and then you have a list of skills right. So you can actually check out the browse list and see a list of available skills. So we can paste this in a second and see what we got here.

Then we have cloud functions as well. So I want to see the available skills that we have with browse.sh and then it will just give us a list here. So layer one is a conductor. That is you.

That was me just using Hermes app. Layer two is Hermes agent right. So this is your single command window. So you type in plain English, Hermes turns your words into agent instructions and calls the right skill right and you can see for example it's pulled in all the different stuff that we can do here which is pretty cool and then we also have the orchestra which is the browser skills.

So this gives you 250 pre-built skills and each skill is one violin in your orchestra and then you also have the score right. So the agent OS. So the agent operating system is your sheet music right. Your SAPs, your memory, your routing, your workflows.

It tells every agent what order to play in, what to remember and where to send the results right. So you can see for example we can switch between our tabs. We've got our goals. We've got our SEO tools here.

We have our studio for generating speech and images and videos. We have the notebook where we're actually connected to our notebook LM where we can pull in for example podcasts. So we have our videos generated here, infographics etc. We have the Kanban board with all of our teams of agents.

We have the journal, the memory and even the build guide which you can see right here right and so everything is set up and ready to go and we've got all our history with Hermes agent as well and then we have the encore which is the compound output right. So the output keeps playing whilst you sleep. This runs 24-7 if you want it to. So reports can get written, leads can get researched, listings can get researched as well, content can go out and you wake up and the orchestrator has performed all night right and you might be saying okay why does this framework actually matter.

Most people stop at layer two which is just using the agent but there's so much more you can do here. 99% of people are behind because they just use one chat window, one tool, one tab at a time right. The conductor stack adds layers three, four and five and that's where you get the most out of this right. The browser skills, the score, the encore etc right.

So one agent is really just a toy but a conducted orchestra of agents is a business and that's the biggest difference between all this stuff right. Now at this point you might be saying okay most people never stop because they believe the wrong thing. So I want to just break down the six biggest myths right here right. So some people want to say I have to be a coder or I have to be technical to set up AI agents.

Hermes installs in 60 seconds right. You can type and play in English, if you can text your friend you can run an AI agent. Other people say I'll end up with 10 different apps and 50 different browser tabs. That's why you need the agent operating system right.

So the agent operating system gives you one command tower, every agent reports to one screen, one inbox, one log. Let me show you an example. So if we say okay go to google.com for me. So you can see it running right here inside Hermes agent and then you can see it's navigating here.

It says navigate to google.com, page loaded successfully blah blah blah right. And so it can navigate the web for us directly which is pretty cool. Obviously people say okay web automation is fragile, always breaks when a site updates or something like that. By the way you can see the agent is navigated there.

So it says navigate to google.com ready for the next action. So we can get these browser agents to just go off on the web and do whatever we want. Browse.sh, we were talking about some people say web automation is fragile and it breaks a lot right. Browse.sh is a public catalog so when a site changes or anything's updated the skill is updated by the community and your agent just keeps working which is great.

Also other people say only big tech teams can run agents 24 7. Honestly like in 2026 a solo operator with a laptop now runs more agents in a day than a 50 person team did in 2020. I would say AI tools just write text, they can't actually do things. So these browser agents can book flights, they can track packages, they can check competitors, they can post content, they can audit websites, they can fill out forms right.

They act, they don't just talk. And other people say if I learn this now it'll be updated in a month. So the orchestra changes but the conductor skill improves right. So learning to conduct agents is really like one of the most powerful tools you can learn right now.

Learning how to manage and build agents is really powerful and I think the people who win over the next 10 years are not the ones who write the best prompts, they are the ones who build the best systems for the AI agents to run inside and that's the biggest difference here. So how do these three pieces fit together? It's a simple picture. So you see that at the top you talk to Hermes and Hermes pulls the right browser skill from Browse.sh, your orchestra right.

So AgentOS handles the routing, the memory, the SAPs etc and the agent just goes off and does the job and the result comes back to your screen like you can see over here and so that's basically how this process works. Now if you actually want to type it out and set it up you can see the instructions right here. We've got a full guide inside the AI Profit Boarding but it's just one line in and then you get a full plan out and that's basically it. Other people say this stuff sounds hard and complicated and using AI tools it seems pretty difficult.

We actually have over 159 pages of testimonials and wins inside the AI Profit Boarding of people just getting all sorts of wins who have never used AI before right. So many people are just getting awesome results just by following the stuff inside there. So it's a great community but you don't have to be an expert in AI. Like the whole point of this is like AI has to be at a point where anyone on the planet can use it and understand it right and so that's why this stuff is getting easier and easier to use.

That's why AI agents are 10 times easier to build today than they were one or two years ago and that's what it's all about. So just to recap here's everything you just learned compressed. You are the conductor, you set the vision, agents do the playing. Hermes is how you conduct the orchestra so it's just one chat window playing English and it calls the right skills automatically.

Browse SH is the actual orchestra right so you get 250 ready-made browser skills with one CLI to command it and then you've got agent operating system which is the score right so SOPs, memory, schedules, routing all in one place. Agents will work 24 7 for you which is great we've broken all the beliefs here so we've talked about how powerful and easy this is to use and if you want my full guide for setting this up along with everything else you can get it inside the AI Profit Board link in the comments description or go to the AI Profit Board dot com you can get the full Hermes and agent operating system setup guide or browser use agents right here. It's a full setup guide as you can see along with all of my new training so we have new video tutorials and step-by-step guides as you can see right here including how I set up the agent operating system with prompts and files to get that set up so feel free to get it. Also inside the community you can ask questions, get help support, you can for example I create video tutorials for members inside there every single day and then also we have four weekly coaching calls where you can ask questions, get help and support live.

On the map you can connect with people in your local city using this stuff. Inside the classroom you can get a beginner to expert or AI automation course here plus all my new trainings like you can see that's all inside here. Let's see what questions we've got now so we've got a question here says what would you say are these top three functions agents can serve for small businesses? The way that would look at that is try and figure out okay what are you spending your time on right so you really want to figure out okay where are you spending your time and then how can you automate that with AI agents?

I'll give you an example so previously I used to spend a lot of time creating content now we have a system for AI avatars and so that saves me a lot of time and my team can just handle it right and so we we actually built a system for that over here so the way that you want to really figure this out is look at okay what do I spend my time on all day and then how can I automate that right so I gave you the AI avatar example another one for example for me previously was that I used to spend a lot of time building landing pages right whereas now I can just automate that with this guide we can easily design it's beautiful it's easy to set up it's nicely organized better than I could organize and design it myself and I did that using my AI agents with Hermes right and so when you set up systems like this they're really powerful but again it comes down to what's your constraint and everyone's use case is different everyone's personal use case is different the cool thing is if I'm like okay my constraint is videos I can just type that in the search bar here and get training on it right away right if my constraint for example is SEO then I'm like okay I can just type in SEO and get access to all of the best trainings on that right here right if my constraint is for example landing pages I can just type that in so and get training on it as you can see right and so whatever the constraint is we actually have training and tutorials on it do you learn to use these tools before creating these videos it's a mix honestly like the way that I try and do it is I try and do a test run before I go on and explain how to do it but sometimes I often say this in the videos like we'll test this out today and just see if it actually works right but also I'm at the point where I've spent so much time building AI agents like I know the process for getting stuff done right I know the process for just getting almost anything set up so that's the way that I look at it is okay I've got so much experience doing this stuff that I know how to get these skills working on the spot