Claude + Hermes Agent: NEW Agent OS is INSANE!

So today we're going to be looking at how to build your own agent operating system and some of the best questions I've had about it recently because I know if those people have questions about this, it's going to help you too because you probably have similar questions about how to create an amazing agent operating system where your teams of agents can work together building amazing stuff and just creating awesome stuff. I mean, for example, some amazing things that we've built recently, like for example, what we actually found was that when Fable 5 got taken down, Fusion just came out, right? And Fusion is a new way to have like a team of agents working together that can build something better on benchmarks than Fable 5. And I've actually tested this out versus Opus 5 and from what I've seen, it actually outperforms Opus 4.8 by a long way. And so this is really interesting. And what we're going to do today is just answer some of the best questions we have about this stuff, how it works, etc. How to set up Obsidian properly, how to get your Hermes agents working with Claude, how to build best agent operating system possible. So let's get straight into the questions. So one of the first questions we had from Omar is, is it possible for an AI system to take ownership of a single long term task over an extended period? You know, so for example, like six months, like how could you assign it the task of improving a website's SEO performance? So, you know, are there any tools or platforms designed for this type of long running autonomous workflow? So one thing that we have inside our agents is we have the power to build out scheduled tasks. So for example, if you wanted to give Hermes agent the responsibility of organizing and doing something every week for your SEO website, then what you can do is just create a task here. Or you can actually go direct to the chat and ask Hermes agent or whichever agent you use to do this for you. And then you can manage everything in one place. If you just go to the Chrome job section inside manage here. And you can do the same with OpenClaw or any other agent. So how does this work step by step? So you would assign a scheduled task to your agents. Now, I would recommend that you either use Claude or Hermes for this for SEO optimization. You would tell it the workflow and tell it every time you do this workflow, you improve it and you look at what's working and what's not working. And you iterate based on that and you update your workflow and then also run it on a daily schedule. So, for example, it could be daily optimization of meta tags. It could be uploading one new blog per day. And that's how you can give your agents responsibility for one task because it's running on a scheduled system. And then also you've got the manage section here with scheduled tasks. So you can easily see or delete any stuff that you need. And to assign a task, you just go to the chat with your agent and assign it directly there. So Andre has a question here, which is Hermes is failing to use Claude code for coding work consistently. So when it actually comes to coding work, I typically prefer to use Claude code in the first place. So for example, actually like building out the agent operating system on the backend, we'll use Claude desktop because it just tends to do coding workflows better. Whereas Hermes is great for just doing little tasks or scheduled tasks or recurring tasks. Now, I'd also question which model you're plugging into that too. So there's two aspects to this. Number one, I would actually use Claude for coding work instead of Hermes simply because Hermes is great for scheduled tasks or admin tasks or smaller tasks. But for coding, I tend to find I get better results with Claude. So when I'm building out my agent operating system, we actually use Claude desktop on the backend. And then when it comes to following instructions and skills properly, I'd also look at which API you've plugged into Hermes. And if it's not doing what you want, I would look at using something else instead. So for example, you could switch to Minimax M3, you could switch to Grok, etc. Quite often, if Hermes is not working that well, it can be down to the brain that's plugged into the agent as well. So Carl was posted about his current build for his marketing agency. Basically, he's got a current setup with he's got like Jarvis as a dispatcher. Sub agents are created from there with a team of advisors and specialists. And he's got a knowledge base with Obsidian now too. But the hard part he's saying is not can AI do the task? You know, it's pretty good at that. But the problem is getting reliable handoffs, clean source rules, and enough QA on it as well. So how can you build better systems throughout quality control, basically? So this is actually something that we've been testing out lately to get better results. And it's working really nicely. So if we go over to the Kanban board here, I typically in big tasks, like let's say, for example, you're trying to create a fully edited video with AI and just, you know, having minimal input, but also making sure that everything that you do is set up nicely together. What we actually do is we have a judge on the team. And so, for example, we created this separate profile. It's called Video Judge for videos. And it looks at the video, sees if it's actually good or not. And if it's not good enough, then it will ask the team to keep iterating until it finally is. So you can see it actually scores the team on their output and the quality. And it actually outlines the critical issues in FNS. So the way that I've been approaching this is building out a team of agents for quality control. So you would have, for example, if you have a video team of agents creating videos, then you'd have one agent that's a judge and scores the team on the quality of their work. And if it's not good enough, they keep iterating until it is good enough. And that way you have a separate agent that's only job is to critique, review, and quality control your team of agents work. And with Hermes, this can work really well inside the Kanban board too. So you can assign a task inside a Kanban board. It automatically gets delegated to sub-agents. And one of those agents should always be a judge who scores the work and tells the team to keep improving until the work is the quality that you need. And in the end, this looked really nice. If we have a look at the final video here, which we can preview inside of Hermes workspace. So this was the finished product, fully edited, looks really nice, nicely edited, and probably better than, you know, what my team can actually do. So the quality of this was just amazing. Once it finished, it looks great. It had music in the background as well. And again, I didn't touch it. It was just my team working together with a single prompt that I gave them. This is quite an interesting one. So this is from John. And basically what he's done is he's set up like twin agents every time a task is completed inside his agent operating system. And basically what he's doing here is he has, for every task, a free agent and a paid agent. And then he compares the difference in quality between them so he can see whether the paid agent is actually good or not. And that's a really interesting way of using AI that I've never seen before. But it helps you compare everything and just get a benchmark for everything that you're doing versus free models. So I absolutely love that idea. That's a great example of what people are sharing inside the AI Profit Boardroom in terms of like, you know, how can you check what your agents are doing? How can you check the quality of them? How can you reduce the cost of everything you're doing as well? So this is a good question from James. James is saying he's got an open clause setup with a principal agent and 25 registered agents with their own workspace, skills, departments, etc. And he's wondering if there's a better way to manage your agents with proper delegation. So, for example, could you have a Kanban board in a control panel? Could you move to Hermes? Should you use a paperclip, etc.? So, also, should you add more knowledge with Obsidian to avoid duplication and add better context to your agents? These are great questions. So here's the way that I would look at it. If you want teams of agents, I don't think OpenClaw is particularly good at that. Simply because it's not really designed for that. It's designed for just having one OpenClaw instance and working together. I do think that Hermes is really optimized for teams of agents because they've created a whole Kanban board and it's kind of native to the Hermes setup. So if you want to control your agents and organize them and you're only sticking to one agent, Hermes is great for this because you have the Kanban board natively. And you don't really get that with OpenClaw. Now, I'm sure there are, I have seen alternatives, but they're just not that smooth when it comes to using OpenClaw inside a Kanban board. Whereas Hermes is native to this. This is what it's designed for. It's designed for teams of agents, teams of profiles that can all work together on a task. As you can see right here when we're creating all sorts of different tasks between our agents. And we can have multiple Kanban boards as well for separate projects to keep everything separated too. Now, when it comes to Paperclip, Paperclip is great if you need multiple teams of multiple agents. So I would recommend Paperclip if, for example, you need different types of agents working together. So if you have OpenClaw and then Clawed and then Hermes and you want them all inside one team, that's where you would use Paperclip instead. And bear in mind, Paperclip is designed for really creating like a company, a team of agents. Whereas a Kanban board is not really company focused. It doesn't have an org chart. It doesn't have a list of issues, et cetera. And also Paperclip is a lot more visual when you use it. Bear in mind as well, if you're switching from OpenClaw to Hermes, they actually have a migrate option as well on the onboarding, which should probably help you. Now, the final one, you know, should you add a memory to your agents? I would say this absolutely improves everything that you do. Because when you have a memory and when you have context for your agents, not only do these teams of agents, for example, with Hermes Kanban boards work better, but also they can automatically update your second brain so that anything else you do in the future is all linked together as well. But you're not updating it. You're not manually improving it yourself. So, for example, everything that we've done here, like even 18 minutes ago, we've got the full daily briefing conversation between myself and Jarvis. I didn't create that. My agents created that. And so when it's automatically updated like this, it's super powerful because everything is easy to link together. And also you're building something. You're building a system that you can come back to over and over again. So I definitely recommend using Obsidian with your agents too. It's great for context. You can get your agents to organize and update it for you so you don't have to spend any time organizing it. And also it just helps you visualize everything much better and link everything better together. And the more you use it every day, it improves and gets better and better as a system. So definitely recommend that. Great questions from James there. This is a good question from Jason who was asking, you know, if you want the infinite context engine, if you don't know what that is, don't worry. I'll explain in a second. How do you use that on Windows? So if you're not sure what the infinite context engine is, this is a system where I have OMI basically records my screen. It records my microphone. You don't have to do that, but this is just what makes it easier for me. And then I export those notes daily to Obsidian. So we were talking about our agents updating our context daily. So this can take notes from me all day. And then what it can do is it can plug that into our Obsidian memory vault. It syncs automatically. So then if we go to Obsidian, Obsidian is updated daily inside our knowledge graph. And that plugs into all our AI agents. And so it's a really powerful way to get a lot of good context and good memories for your AI agents. Now, if you're on Windows, the only way that I know, the only way that I know to use OMI on Windows is if you have a wearable. Because it doesn't seem to work natively with Windows. You could have it on your phone instead. There's also this that I've set up previously. And this kind of does a similar thing to OMI. It's an open source project. You can see it's got 20,000 stars on GitHub. And that's a pretty interesting project as well. So that's something else I would look at. And it does work on Windows. So you can see here that it's designed for Windows too. We actually have a full training on how to use it. And again, you could take the context from that and then plug it into Obsidian instead. That might be a better option. Now, this is pretty interesting as well. So Benjamin was sharing a voice agent that is created inside his agent operating system. And this is kind of a take on Hermes Jarvis, which you can see over here. So we can voice activate our AI agents, speak to them with a chat. And then we can see the whole conversation history here. Plus everything that's built inside our agent operating system too. And that's pretty amazing when you think about it. Now, Benjamin has built out the Logos Oracle. And so you can speak to any philosopher that you want, right? It's trained on philosophy, which is pretty amazing in itself. So you can see here, like it's not just for me, like anyone who's even non-technical can build with this stuff and you can get really creative. That's one of the best things about using, for example, an agent operating system instead of using like Hermes desktop or these other apps. The reason for this is like when you do this, it's really powerful because you can customize it exactly how you want it. And that doesn't have to be for business, right? It could be for personal, as you can see with Benjamin here. So it's a fantastic example of what you can do and how you can customize things exactly how you want it. So love the idea, love the creativity, super impressive. So that's basically it for all the questions. I've answered everything and all the questions I've had recently about using agent operating systems inside the mission control. If you want this whole setup, by the way, you can get it from the AI Profit Boarding. Link in the comments description or go to the AI Profit Boardroom. You get the mission control, you get the paperclip set up, you get the AI agent mastermind where agents can actually talk together and you can see all the chat history too. You've got Claude inside there, Hermes working, you've got Hermes Jarvis, the studio, you have the Kanban board system, as you can see. We've even plugged in Fusion, which is like basically Fable 5 level intelligence. But, you know, you don't have to wait for Fable 5 to come back out, right? You get something even more powerful. So this is something super interesting that we've been building out and testing and it's creating amazing stuff, as you can see. We also have GLM 5.2 plugged in there, Kimi Code. And the great thing about this is as well, we've plugged the memory into here as well. So you can use Obsidian with your AI agents too. So if you want to get that, that's all inside the AI Profit Boarding. Link in the comments description or go to AIProfit Boarding.com. This is my AI automation community that helps you scale, learn and save time with AI automation. We've already helped 3,600 business owners who are active inside there. We've got over 182 pages of wins and testimonials from active community members. So I know like if we can all get results, you can get results too. And it's just an amazing community where we learn and we grow together. And I really care about the results, you know, of everyone inside there. That's why I answered these questions personally every day. Now inside the community, you can ask questions. There's always people online, which means you can get help and support. I personally answer these questions too. Inside the classroom, you can get access to all of my best trainings and guides, including the agent operating system, which you can see here. And we update this daily. You get video tutorial, plus you get the installation zip file. So you can just quickly install that. And then we add new daily trainings based on what's actually useful. And I test all this stuff personally. Inside the calendar, you can jump on week coaching calls, get help and support in real time. And inside the map, you can actually meet people in your local area who are using AI agents just like you. So it's a fantastic way to just meet cool people, make lots of friends, and also meet people who are doing similar things to you, right? So feel free to get that link in the comment description or go to the AIprofitborn.com. Thanks for watching.