AI News Today
← All episodes
Episode 18 · June 3, 2026 · 10:56

NEW Hermes Agent OS Just Changed AI Forever!

Why Hermes Needs an Agent Operating System (Mission Control Dashboard Demo)

The script argues that using Hermes Agent in a terminal or even Hermes Desktop limits its value because it becomes a single-chat experience without strong organization, artifact previews, or team/agent management. The speaker showcases a custom-built “agent operating system” as a mission-control dashboard that centralizes multiple agents and tools, previews outputs like websites and media, saves everything in one workspace, and enables autonomous “goals mode” for long-running tasks. They highlight faster access to features (like voice mode), the ability to quickly plug in new releases (e.g., Minimax M3, Claude’s Ant CLI, MCPs), and a memory system via an Obsidian vault to avoid repeatedly re-explaining context. The episode concludes by promoting the AI Profit Boardroom as a place to get the system, tutorials, updates, and community support.

00:00 Why Agents Need Systems
00:42 Terminal Use Is Limiting
01:03 Mission Control Dashboard Demo
02:03 Hermes Desktop Problems
03:22 Voice Mode Instant Setup
03:58 Plug In New Tools Fast
05:31 Memory With Obsidian Vault
06:47 Kanban Made Simple
07:35 Build Systems Not Models
08:52 Get the Operating System
09:19 Community Proof Anyone Can
10:10 Boardroom Features and Wrap

Full transcript

This new agent operating system for Hermes is absolutely wild. I'm going to show you exactly what I built, but this is basically what you need with your Hermes agent if you really want to get the full power of it. And the reason that I say that is because the way that I see people using AI agents, and you might have experienced this yourself, is like nobody knows the actual use case. Nobody knows why they're using it because they don't have a good system to manage their Hermes agent or any sort of agent in general.

And the problem with that is that they don't really get the most out of it. Now, bear in mind, this is the most powerful technology we've probably ever had as human beings, right? And it's totally changing the face of the planet. But the biggest issue is nobody really understands it.

And the reason for that is if you're using Hermes agent, let me explain. If you're using Hermes agent normally, you're going to be using Hermes inside the terminal like so, and then you've got this terminal window and you're just operating on a single chat basis. You plug in a prompter, you get something back, but that's all. And that really is just a glorified chat GPT if you're using Hermes agent.

There's no point you even using Hermes agent at that point. You're just wasting your time. Now, what I think, and this is what's worked for me and totally changed everything in terms of the way I manage my AI agents, is having one dashboard, one mission control with all your agents built in. And every time something new comes out that's actually useful with, for example, Hermes, you can plug it in.

Let me show you an example here. For example, Minimax M3 comes out, probably the most autonomous agentic model that's ever been released. It can run for 24 hours on its own. So what do we build in?

We build in a goals mode into our agent operating system. We plug in a goal and then it can go off for 12 hours, 24 hours. I can go to sleep, come back, and it's getting work done. Plus it's all saved inside my workspace.

This is another powerful use case for an agent operating system is that you can actually save everything in one place and preview everything that you've created. So here's some examples of what we've built, right? And you see the quality of the websites as well. It looks super nice.

It can even, for example, do like trailers, voiceovers, podcasts, videos, and everything is in one place. Now tell me, how are you going to find that if you're using the terminal? You're not, right? Now you might say, okay, Julian Permy's desktop has just come out.

Let's do that instead. But the thing is, you're going to have the same problem, which is where are your other agents? Where's your team? Where's Claude?

And also, how do you preview your artifacts? So as an example of this, what are you looking at here? How's it going to help you? So for example, if we click on this markdown file that we've created, we can't really preview it inside the app, right?

You're not going to get a full preview. It's not going to help you. You can't really see any images you've previously created in Hermes desktop unless you start creating them in the future, right? Whereas for example, you compare that to an agent operating system and it's just a mission control.

You are the captain of the ship. You are the one managing everything, keeping everything in one place. And you do it in a beautiful design that you can custom make anytime you want. That's the thing.

I have made this. This is something that I've built from scratch and it's customized exactly how I want it to be. The reason it looks like this is because I've decided it looks like this, right? And it's the same for you.

You, as a business owner, probably one of the reasons you got into business is because you want control and also you want freedom. You're not going to get that if you're using the terminal, right? Unless you're a coder and then maybe you have a bit more freedom. You have some skills and some chops with Hermes agent that you can use inside the terminal.

But most of the time, 99% of the time, this is going to be something that gives you way more power. That's going to make you feel way more in control of the agents that's going to help you get the most out of them, right? And the other thing I would say here as well is that if you're using Hermes directly inside the desktop, quite a lot of the things that should be working aren't working. I'll give you an example.

So if we want to do voice mode here, we have to allow a microphone, but there's no option to allow my microphone, right? So how do I do it? I have to figure it out, spend 20 minutes Googling it or going inside chat GPT or Claude or whatever. Whereas for example, if I go inside Hermes on my agent operating system, I can just click this button, boom shakalaka, we can start having a conversation.

How are you today Hermes? Ah, boom shakalaka, right back at you. I'm great, thanks for asking. Ready to chat whenever you are.

You see the difference right there? That is what we're talking about. Now, also one of the big things that I would say is like whether it's with Hermes, whether it's with Claude, whether it's with OpenClaude, you can switch between them. You can create and organize things like you see here.

You can add new features in the functionality. But also the reason I'd recommend organizing your agents with an operating system is that you can add new features. So Claude just dropped this new feature called Ant CLI and it came out about five hours ago. And guess what we did?

We plugged it straight in as a CLI into the agent operating system that we have. And the great thing about that is that it makes it 10 times easier to see and organize and add new things in. So we can take advantage of that new system. Whereas for example, if I was trying to figure out Ant CLI on my own, what a nightmare that would be because I'd have to look at the update here.

I'd scroll down the page, looking at all the demo videos, a bit overwhelmed. And then I click on the API documentation. And if you're not a coder, tell me how you're going to understand that. Tell me how you're going to use that, right?

It's going to be very difficult and very time consuming. It's the same with news research. Like when they released, for example, Hermes with MCPs in the MCP catalog, most people didn't use that because they didn't understand how to use it. Whereas for example, if you plug into an agent operating system, this is a beautiful thing about CLIs and also stuff like Hermes with the open source projects is you're like, this sounds cool.

I don't know how to use it, but let's plug it into the agent operating system so that I actually can use it in a very quick way. Let's say something comes out, 10 minutes later, boom, you can plug it in. And so this is why I recommend using Hermes with an agent operating system and also how powerful this is. This stuff is crazy, right?

The other thing that I would say here is like, if you've got, for example, lots of stuff going on, which I'm sure you probably do, then you probably want a memory system, right? Now, how do you get a good memory system set up? We have an Obsidian Vault, which you can see here. All these are different things about me and different notes and that sort of thing.

So you can see that we can plug in and look at these memories and everything that we've created. And then my AI agents can plug in the dots and use those to get good information. So if we go back over here, you can see, for example, we have all of the memories inside our vault that my AI agents can pull from. Why is that useful?

It's useful because it helps my Hermes agent take that information, understand how it can apply all of that context. And then also it saves me time and energy because I don't have to re-explain myself about 100 times, which I normally do if I'm having to go into Claude or chat cheaply and give it context. So for memory, for context, for UI, for speed of use, for efficiency, and also just for breaking super technical stuff down and turning it into something that's actually useful, basically being the bridge between new updates and actually implementing those updates, I would recommend an agent operating system. The other cool thing I like is just in general being able to switch between the sessions and the skills and the plugins and seeing everything.

And then the final thing I would say, and this is particularly important for Hermes agent, is that having an agent operating system with a Kanban you can control is great. If you wanted to use a Kanban normally, let me show you how to do that. So you'd have to go into Hermes News Research and look through this tutorial, which is, again, I think in the modern age, most people are not reading blog posts. Most people are not going to spend an hour trying to understand all the terminal commands and everything else.

So how do you condense that into something you can actually use? How do you take this technical information and make it implementable? What you can do is you just, for example, you set something like this up, where it's like, okay, I just have to type something in and then we get it running, right? So I type in build an AICO keyword research strategy, boom, we have that built in.

That's what you want. That's how easy this stuff can be if you have the right systems in place. And that's the final thing that I would say is you don't want an agent. Agents come and go, right?

You don't want an API. APIs come and go. You don't want a new model. Models come and go.

What you want is a system. You want a system where you can organize everything. And this is something I've learned from running my own business is like, when you've got, I'll give you an example. So if you look at Google and the way that Google has become so successful as a company, it's because they have amazing systems.

The people, they come and go, right? The CEOs and the directors and the teams and the amazing talent that they actually hire, they will come and go over the years. But the system, aka, for example, just going into Google, this system has been there since day one and it will be there for probably the rest of the history of Google, right? And so why is that?

It's because you need a system and that's how powerful agent-operated systems are. That's why they're so useful. If Cloud Mythos comes out next month, no problem. We plug it in.

If Hermes Agent releases version one instead of v0.15, no problem, you plug it in. So everything is interchangeable and that's why you want it as well. And then also you're gonna have a positive feedback loop. Whereas for example, you test some of this stuff out and you're like, there's something here that I don't like.

No problem. Let's tweak it, make it better for the next time, right? So you can have this positive feedback loop where the more you try it, the better it gets because you're giving it more feedback on how to improve it. And so it gets better and better.

It's this positive cycle that goes up. So thanks so much for watching. If you want to learn more about building an agent-operated system, you can actually grab mine from the AI Profit Boarding. Link in the comments description or just go to the AIProfitBoarding.com.

So you have the agent-operated system right here. We have a video tutorial on how to use it. And not just that, we have updated daily zip files for all the new features that we build in. So every time we build in, the new last updated date, we give you a video tutorial on how to use it.

We have a community that's building with this stuff and you can get the zip file right here. And the one thing I would say about this is that some people are gonna look at all of this and think it's too difficult for me, too technical for me. It's not. If you can message a friend, you can use AI agents.

If you can text a friend, then you can build a mission control or you can just grab mine and install it. But you can see, for example, Mike, he's from the AI Profit Boarding. He's got his mission control running and he said it's absolutely amazing, right? You can see, for example, Ravit as well, built out an agent operating system, integrated it with NA10.

And now he just has it command and it builds the automations for them, right? Same for David. David built out a beautiful agent operating system too. My point here is that anyone can do this.

I've seen so many people who've never touched AI get amazing results with AI when they have the right systems in place. So thanks so much for watching. If you wanna get the system, get it inside the AI Profit Boarding, link in the comments description and go to the AIProfitBoarding.com. We have an amazing community of 3,300 business owners where you can ask questions, get help and support, et cetera.

Inside the classroom, you can get all of my best new tutorials. So we actually update this daily with new trainings, as you can see. Every time something new comes out that's actually useful, we will break it down into a full step-by-step guide. As you can see, with a 30-day roadmap, 100 prompts for implementing it in a video tutorial.

Plus you can connect with me personally. We have weekly coaching calls. We have four-week coaching calls where you can jump on a call, ask questions in real time, get help and support, meet other cool people as well. Inside the map, you can actually connect with people in your local city who are building AI agents like you.

So you can see that across every major city, we have people all over the world building out and you can connect with them personally too. And you can get that all inside the AI Profit Boarding. Thanks for watching. See you inside there.

Cheers, bye-bye.

More episodes

Browse all episodes →