This Claude Agent OS Is INSANE!

Today I want to talk you through my Claude agent operating system, what I'm doing, what I'm automating right now, how I use Claude and why I've decided to put it into a whole system plus some of the new stuff that's come out literally today that I just changed around and we've plugged directly into this system right here. Now, if you're wondering, okay, what is an agent operating system? Basically, it's just a mission control for your AI agents so that we can plug everything together. Why would we want to do that?

Because we can keep everything organized. Why do we want to do that? Because for me, and I don't know about you, but there's no way I would be opening up chat GVT and then Claude and then navigating somehow to super grok and then going over to all the other tools that I use and just messing around like that. Right.

Instead, what we like to do is have everything in one place. So it's like, okay, where's Hermes? What's right over here, mate. All right.

How do we navigate to open Claude? There we go. Right. How do we create videos?

Well, we're going to just go straight to the studio over here. Right. Well, how do we use that new feature that just come out today? Well, we can just plug in.

Right. And so one of the best things about having an agent operating system is that every single day. Number one, I'm excited to work on it. And number two, we could just upgrade it and we can improve it.

And I'll show you an example of this. So if you haven't seen it already, basically there's a new update that came out today, which is Ant CLI, right? Just came out for the Claude platform. Mintroy actually posted about it today as well.

And basically what this means essentially is that you can use Claude's platform with a terminal. So it connects the whole of Claude's console to a terminal where you can speak to it. And so what does that mean? That essentially means that you've got a CLI where you can talk to your cloud hosted AI agents.

Now, why is that useful? It's useful because if you're using Claude code, well, that doesn't run in the cloud, right? That's just running when you've got your laptop open. And then for example, if you could link that to your managed agents in the cloud, let me show you an example of that with Anthropic.

This is a platform, right? And so if we have AI agents plugged in, which we do, then we can speak to those agents and use those agents whenever we need to. So here's a couple of examples. Now, the way that I've done it is literally this morning, we sat down and we set up four different AI agents, like you can see, and then we can give them different tasks.

So if we go to my SEO strategies over here and we're like, right, okay, link building ideas for a website about AI profit boardroom community, right? Something like that. You can then work in the background and it's just working in the cloud on that new strategy using the new system that we built, right? So we've plugged our AI agents that run in 24 seven, we've plugged them into our agent operating system.

And so that means again, I don't have to like log into console and then click buttons and then navigate and then remember where are my AI agents, right? There's no way I want to mess around with that. And so you can build new stuff like this all the time. And that's one of the best things about this is like, there's no real limit in terms of what you can do with this.

Your only limit is your creativity. Now, on top of that, when you're building out an agent operating system as well, the one thing that I would say, and this is something that I've learned the hard way is just keep it minimal. So you don't want loads of tabs everywhere. You don't want to add every single idea that you have.

You really want to think, okay, how can I keep it lean and useful, right? And the more that you can eliminate the stuff that you don't use, the better it's going to become because it's not bloated with loads of different features you don't use. I'll give you an example. So I've rarely used the ultra code feature.

This is something that just came out last week and it was fun building it out. There's some cool stuff we could do with that. Do I use it every single day? No.

Do I need it? Probably not. Right. And so that's a feature that we could actually remove later.

And then the other thing here as well, is like, because it's connected to your AI agents, I could create an image inside Hermes and then I could take that into Claude. And if I have a conversation with Hermes, that will actually be logged into my Obsidian memory for here. And then what we can do is we can take that memory and plug into Claude later. And so everything kind of links and weaves together effortlessly without you messing around.

So this is a new way that I find is 10 times more powerful, 10 times more efficient if you're using AI agents. And if you're still in the terminal, you know, let's open this up. If I'm still in the terminal and I'm talking to Claude like this, right. It's inefficient because number one, I have to type Claude to get access.

Whereas I can just navigate to this website, which is, you know, I think at this point as well, if you look at the way things are going, everyone's just using a Chrome machine, right? It's not really a laptop. It's not like you use apps or do anything outside of Chrome. For most people, for 99% of people, you know, they're using Google docs inside Chrome and then they're switching to something else.

And so if you can keep everything inside one place, again, you're going to minimize the amount of time. So you're like clicking around doing different things. And also it's just a lighter cognitive load. If I'm, for example, thinking about, okay, how do I go from Google docs and then switching over to terminal?

And then where do I find that terminal tab ad with Hermes doing something? It's going to be super, super time consuming. And just like, you know, I think one of the worst things you can do when you're working is have like a heavy cognitive load where you're context switching all the time. You know, in the same way that, you know, you see with AI agents, well, if you're switching between contexts with your AI agents, the problem is that you will burn through tokens and it gets quite expensive on API cost if you're switching between projects all the time and different tabs and everything else.

And I think it's the same for a human, right? I think the human brain works in the same way where it's like, you know, if I just have everything in one place, isn't that 10 times easier to manage than having everything inside terminal and then cloud and then cloud desktop and then having a window that I can't find anywhere else. So just to recap here, we've talked about the ANT CLI that just came out and how we're using that with managed agents running cloud, and we can give them scheduled tasks so we can say, okay, go off and do this every day and we don't need to be able to code, I don't have to click buttons to set those agents up. I just go inside the ANT CLI and then we can plug that into our console.

Pretty powerful. Now we do have this ultra code thing. I think this is really interesting if you need something where you can have a huge team of agents and you're just tackling big problems all the time. Let me give you an example there.

So let's say for example, if I'm trying to like mass scale SEO content and I'm like, right, okay, I need something that just works autonomously as a dynamic workflow with loads of different sub-agents to optimize a website and create like 150 pages for it, a beautifully organized and all optimized for SEO. Okay. That could be good. But how often do I do that with ultra code?

Not that much. So I just want to give you an example of what you could do with it. Do I use it? Not that much.

And then the final thing, and I absolutely love this, this has saved me so much time, is that when we're using the workspace inside cloud, I can come back to everything that I've previously done. And that is very, very useful. So even if, for example, you are using the cloud code desktop, like you see right here, how do you find everything you've created? You can't, right?

And then also, if you are using the terminal, how are you going to find everything you created? Good luck with that, right? You can't even preview it, let alone see your previous projects. And so if you could have all of your workspace neatly organized into one amazing project, like you can see, isn't that 10 times better for sure, right?

And then we can see everything that we've created previously. We can switch between projects. We can switch between agents. It's just much easier and nicer to organize, right?

And you can see some of the stuff that we've built with this, just playing around with it and that sort of thing. So I like that because then I can come back to the projects. And if I've coded something like last month or last week, I don't need to worry about where can I find it again? So I think these are all great ideas.

So if you're watching this and you use cloud code a lot, and you want some ideas on how can you organize everything into a neat, tidy workspace, this is it. This is the future. This is what 99% of people are not doing yet, but they should be. I've seen loads of people inside the airprofit boardroom, for example, and everyone is building out agent operating systems.

You can see, for example, Mike, he got his mission control up and running. You've got Rabbit who also did that. You've got, for example, David created one. Loads of people are doing this right now because I think it's just a huge mindset shift that I'm seeing with like the people who are basically early adopters of this stuff.

And I always try and look around me and see, okay, well in my circle or in my community or out of people I know, we're usually early adopters of technology, particularly for AI as a business owner, because we need to be. So what are we doing right now? Because that's the trend that's going to happen in five or 10 years time. And I think this is the way that it's going.

It's like, you've got this mission control dashboard of everything organized. And then also the final thing that I would say is like, I get Claude to fix a lot of stuff in the background. So if something breaks with openClaude, something breaks with Hermes, typically my go-to is using Claude and I can just go inside here and say, okay, fix this locally, Hermes is not working, or I need to add this into the agent operating system, go and do that, and it will just organize it for me. And then finally, you know, I, again, I don't really use the goals and the journal section of this agent operating system that much, I do use a video and the SEO agents like daily.

So that just gives you an idea of how I'm using agent operating systems, why I'd recommend it to you and how this all comes together as one for Claude. And then the other cool thing is if Claude releases a new model, we just plug it in. If Claude releases a new feature, we plug it in, right? And then you can cut and you can add more stuff depending on what you're using.

And also what the people around you are using as well. That's something that I pay attention to. It's like, what are other people doing this for? What questions am I getting inside the AI Profit Boarding?

And then how can I apply that to everything that I'm doing? Because that kind of gives me a feel for what's working and what's not. So if you want my full agent operating system, you can get that inside the AI Profit Boarding. We actually do a lot of coaching on Claude, loads of people using Claude inside here.

You can drop a weekly coaching course too, so you get four coaching course a week. And if you want the agent operating system with the zip file from today, you can get that inside the AI Profit Boarding. You see when it was last updated, you get a video tutorial on this stuff. You also get new daily video tutorials on AI agents.

This is something that I'm testing all the time. If you want to learn the full Claude Obsidian setup, we've got it there too. And you can also connect with people in your local area who are using AI agents like you as well. So feel free to get that link in the comments description or just go to the AI Profit Boarding dot com.

Thanks for watching.