Hermes Agent OS CHANGED Everything 🤯

Hermes agent operating system just changed everything. I'm about to show you the exact setup that's helping real people automate their entire business. That means lead outreach, content calendars, customer support, done automatically, and you can even do it for free. I'll show you how later.

One of our members actually set this up in under 30 minutes and said it gave him the confidence to finally do the things he used to only dream about. So in this video, I want to show you the exact questions and answers our community asked, and the one thing you absolutely have to set up inside your Hermes agent OS if you want it working for you 24-7. Stick with me to the end, and I promise you, you're going to walk away with a completely new way to save time and get more done every single day. Let's get into it.

So we're going to have a look at some of these questions from the community. We've got a question from Amanda here, which is not really a question, but it's a win. So Amanda says, Hermes agent just changed my business. I have to share this because I was genuinely blown away by what Hermes agent has done for me this week.

So she said, I've been using it to automate tasks that I used to spend hours on, and the results are wild. So for example, automating, these are some really good use cases, by the way. So automating lead outreach, building out content calendars, and customer support. Like, it's absolutely amazing when you think about that and you look at that, you're like, whoa.

It just makes me happy to see people achieving amazing results like that. So shout out to Amanda, absolutely insane. What an inspiration, absolutely amazing, well done. Let's see what else we got here.

Are you running open source models for agent OS? So if you have a look at the systems that we've got here, typically the models themselves, like the actual APIs that plug in to the agent operating system are essentially open so that the agents themselves are open source, right? So Hermes is open source, OpenCore is open source, FreeCore code is open source. These are open source agents, right?

But then when it actually comes to the models inside them, I usually use something like OpenRouter, like our alpha, or for example, with Hermes, I have Grok 4.3 plugged in. So the API, the brain behind it, I will use a closed model, but then for the actual agent itself, I often use open source models. So it's a mix. How is the new release of Gemini Omni?

Let's try it out. So you can see some examples here. Like, it looks really cool. There's all sorts of cool stuff you can do with Gemini Omni as well.

By the way, Gemini Omni is like the new version for creating videos with Gemini. You can see some examples here. Like, it's pretty awesome. I think you can just access this at Gemini.com.

So it's pretty simple to do. Yeah, but it looks good so far. So have a look here. So surf says, I got it.

I figured out what I need. Is there anything I can download to a llama that can act as an agent and do research on the web? Because Hermes cannot do that. Or how can I get Hermes to be able to do that?

So two different things in one there. So you can actually use Firecrawl with Hermes. This allows you to browse the web and you can connect this to Hermes directly. The other option that you have is you can use, for example, Browserbase, which we talked about just now.

And that allows you to connect to the web. And then the other one that I was gonna say, and this is really good, especially if you're trying to find like trending topics, is Grok and Hermes Agent. So you can do Twitter search. And with those three setups, you can basically use Hermes Agent to do web research for you.

So we've got another question from Warwick here. Welcome, Warwick, for joining. So his goal is to automate and elevate the content they produce, right? And so one of the biggest roadblocks is LinkedIn.

He wants to be able to generate content, create the posts, et cetera. So I'm gonna show you what works for us on LinkedIn. Here's an example. So if we go over to LinkedIn here, you can see my impressions are pretty good, right?

We have a good run on LinkedIn. We get a decent amount of reach. Sometimes we reach peak of 21,000 people in one day, which is pretty nice. So how do we do this and what we do?

So if we have a look here, and we look at my recent posts and that sort of thing, you can see some examples. And basically, there's two things that we do. And we have the same content strategy that we use on LinkedIn, we actually use on Twitter as well. So we actually, for example, we'll post these AI avatar videos.

They seem to do pretty well on X and LinkedIn. And then also the other option here is just kind of like caption posts. So let me show you an example of that. If we go to X here, you can see some examples.

So here's a post that we recently created. And basically, you can see like, we had a video or some sort of media here. And then we have the full post here, right? With the captions and everything else.

If you want to automate that, we have the full prompts over here. Let's see what other questions we got. So would Hermes be considered an agent harness? Yeah, that's the way that I look at it.

It's a way of basically taking an API, plugging it into something that just makes it really powerful, right? And the better, it's kind of like Hermes is hands and the API you plug in is the brain, right? And so the better the hands you have, aka Hermes, the better the outputs you're gonna get. Omni is supposed to be the nanobanana video.

Yeah, it seems pretty cool so far. And then Inspire Michelle says, sorry, Julian, the graphs are a bit all working for your training, but this screen you have Mission Control. How did you get the agentic operating system? You talk about Hermes, is it like Claude?

Yeah, so you basically, you take the setup for Hermes and you plug it into the operating system. If you need the training on that, go to the new daily update section here and then go to the Hermes and Claude operating system here, right? And we have the guides, we have the full setup here, all the prompts, and then we also have the zip file that you can use, like you can see down here, ready to go for you. Fabricant says, hi everyone, good to see you.

Michelle says, I'd like to get more leads. Yeah, so honestly, I think if you need more leads, then you just create more content, right? And if you need more help on those systems, you can see what works for me inside this section, right? So you can see, for example, we use these channels here.

These are two of my favorite lead generation methods and those systems work really well. So I'd recommend those. And then ClickSang says, do you have any idea of building public works? 100%, right?

If you see, for example, people, this is basically what I do, right? Every day I document the journey, the process, what I'm building, et cetera, works really well. Works really well. I have a lot of friends who do that too.

It works really well for them too. So yeah, I'd recommend that for sure. It's a great way to just be authentic and also share your journey, which is what everyone buys into at the end of the day. Let's see what other questions we've got here.

So Egbert says, are there any small, medium business owners that want to share their adventures, et cetera? So that's really what we do inside this community. So feel free to share anything, post questions, share anything that's worth for you. You can always post inside the community and also inside the classroom and share everything that works for me.

Bear in mind, like I'm also an SME and I share everything that works for me inside the classroom. With all the tutorials and everything else. And then also if you go into the win section of the community, you can also see loads of cool stuff that people have built and shared. So if you go to this section and then wins, you'll find like all the cool stuff people are setting up, which is pretty amazing.

So this is a cool question. So when it comes to AI services, what's the best thing to focus on, right? If you have to focus on one direction, what would it be across all of these? I think for me personally, I would focus on lead generation, right?

I'd focus on lead generation out of everything. The reason for that is like, when you focus on lead generation, that's really what companies care about, right? Like they don't care so much about the automation or which agent you use or which API you use. They care about the end result.

And the most important end result for 99% of businesses is like, what do they get back? For example, how many leads can you generate? And also it's great for you because you can demonstrate, okay, here's what I did and here's how many leads it generated. So you can show like a direct output from it, which is awesome.

Let's see what else we got here. Inspired Michelle says, thank you very much. Your training is amazing. Appreciate it.

Yeah, so you want to join the AI Profit Boarding is where you want to be for all this training that I'm talking you through today. If you want to connect with me personally, just DM me inside the AI Profit Boarding. Can you show us how you could have an AI agent answer questions like you are? Yeah, so like the way that I would do this, it depends, like you actually, it depends if you're looking for like a video agent, one that can answer questions like this, or whether you're looking for a chat agent.

If you let me know, then I can answer that based on your answers. All right, and Tony says, hi there, just joined. Welcome here, Tony. Do you know where I can find the step-by-step for building the agent operating system?

Yeah, so you go inside the classroom here, go to new daily updates, and you can find the agent operating system here. It has the zip files and the prompts inside here, along with a nice video tutorial. Jerry, sharing a win. So Jerry said he set aside two hours a day today to set up his first scheduled task on Hermes, took less than 30 minutes, set up some really cool stuff with Etsy, created a trend report, pretty amazing stuff.

And you can see here, it says the AI Profit Boarding Room is a game changer. It gives me the confidence to finally try things with AI before I just dreamt of doing. So wow, it gives me goosebumps a little bit when I see like people just getting awesome results like that. So number one, shout out to Jerry.

Like what an absolute legend. What an absolute legend. Welcome, and well done to you. It's really cool to see the interesting use cases.

So like taking the ideas that we all share together, and then putting something together that's just so specific and niche to you. Also, I've noticed the same thing, which is like the idea of setting something up is much worse than the actual setup in reality, right? It's always faster and easier than you think. You just have to sit down and get it done.

So congrats, and I feel really inspired reading what you said so I appreciate that. All right, Sachi says, does anyone use their agent for this sort of stuff? I think we might have something like that. Let's have a look.

Oh, nice, look at that. So if you're using it here, you can actually set it up with MCP. I don't know how powerful it is. I would always check it yourself, manually do a test and just see what you think.

Essentially, you could set this up, right? So you can go to Cloud. Let's do this now. So if we go to Cloud here, then we're gonna go to Settings, Connectors, Customize.

From here, you can click on the plus sign, Browse Connectors, and there you go. Look at that, like this. And then from there, what you can do is connect this and it should be able to actually do stuff for you, right? So you can see here that it has a bunch of tools built in and some examples of how you can set this up.

So that could be a really good option there. The only thing I don't know 100% is whether it can actually make changes for you, whether it just pulls in the data from your account, but either way, it could be really cool. All right, we've got another question from Sachi, which is, Julian's latest video style, insane. He's doing three videos a day.

How does he do it? Here's how we do it. Let me pull this in for you. So you have training on this over here.

We actually have training on how to do this here. And the really cool thing about this is you can give it to your team. It's a full system, works really well, and it creates amazing stuff. So definitely recommend checking that out.

For APIs, usually I use a lot of free APIs, right? So it's not really a problem for me, but also it's not really relevant to most people watching because I think being someone who talks about and shows AI every day and keeping up with the latest updates, my usage is gonna be very different than 99% of other people. Oh, this is cool. So Inspire Michelle says, yes, there was a post for someone who posted about you and the work you do.

It was very touching. I think his name was Andrew. Really showing the hard work you do. Appreciation, that's awesome.

I really appreciate that. Yeah, for this, just go back to the side of the video and you'll see it. I have a question in the calendar. So it depends.

The calendar inside the AI Profit Boardroom is for the weekly coaching course, right? So if you wanna jump on those, you can see the times in your calendar time just by going to the calendar section here. That's different to what I'm doing right now. So Sachi says, has anyone had good results with developing an operating system where your agents are constantly working towards your KPIs or quality goals?

I still find a lot of back and forth, keeping them on the chat, spinning, et cetera. The way that I would probably approach that, so I think something like that, you probably have to approach manually, but what you could actually do is create a scheduled task inside Hermes to check in with the goal that you give it, right? So if it's a long-term goal, for example, like ranking a website, then you could have Hermes drop you a note every single day in terms of where it's up to on the goal and the progress made. And that way it proactively comes to you and every day it reports back to you in terms of its progress.

By the way, if you want me to answer your questions like this, feel free to check out the AI Profit Boardroom. That's where I answer this sort of stuff daily in a video tutorial, plus I give you the actionable stuff right here. So for example, if I wanted to set this up inside AgentOS, what I could actually do is I could say to Hermes here, okay, my goal is to, and you could say, okay, my goal is to build a full website that ranks on Google number one for the keyword, blah, blah, blah, right? And then what I can say is schedule a task daily based on my goal, which is, insert goal here, check in with the progress, plus give me a breakdown of all the actionable ways you've actually moved towards that goal, right?

And by doing that, you get a proactive report daily of where you're up to. But if you could put a virtual avatar on the front, that would be impressive. I haven't seen anything that does, like we had an avatar on the website like last year where it could like, for example, accept phone calls and then it could reply to people like a voice agent. That works really well, but I've never seen anything good for live video agent.

I think Pika were working on something, but they released it in beta and not everyone could get access, which made me think, does it work or does it not? So BEYOND is making awesome progress. Shout out to BEYOND. This is pretty interesting.

So Surf asked, is there anything like Manus but free? So you can get free cloud code and that's available to use for free. So that's like Manus. And then the other option you have is you can just use Hermes, right?

So Hermes agent is a free open source project. And that's basically like Manus. The only difference is the, you want a nice user interface with Hermes, right? Which is what you're asking for.

So you can get that with the agent operating system and then you don't have to worry about a terminal window. So like, for example, for me, I don't like using the terminal either, but if I use, for example, Hermes inside the chat like this, it looks way nicer, right? It's way more fun to use. So that's the way they recommend that.

And then when it comes to, for example, other options and you've got cloud code as well, you can use that for free too. So Paul has created Agent Vibes. Shout out to Paul. Let's check it out.

So it's an open source project. As you can see, it's like text to speech for cloud code, which is amazing. So basically your AI agents can talk back to you, which is really cool. So that's an open source project.

Shout out to Paul. He's inside the airport for William and just created that. It's pretty amazing. And he's got a full tutorial on how to use it right here.

So Britton was talking about OMI Chrome extension issues. I actually don't use, basically if you're not familiar with OMI, OMI is a way to build like a knowledge system, right? So you can create a second brain with OMI where it records what you do all day. And then what it can do is export those memories over to Obsidian.

And then Obsidian has some stored inside your agent operating system, like you can see over here, right? And so you get this amazing diagram where everything is linked. You got all of your knowledge together, all the people in your life, everything you work on. And it's like a full amazing sort of knowledge graph of everything and everyone in your life, right?

And so I don't personally use the Chrome extension. I don't think you need it to get results, right? Because if you have the screen recording set up, then it's analyzing what you're doing. And also the same with the microphone, right?

It already has context from what you do there. So for me personally, I don't use it. And also I tend to find with Chrome extensions, they can slow down Chrome a bit too much and then that's no bueno. So Molly says, too much work, the good kind.

Grateful for what I've learned from Julian in this group. Scaling back to focus on what's committed. Yeah, I would say it's always better to focus and do less, but do it better. The way that I look at it is a lot of people, what they do, especially with AI is like they try loads of different tools, different apps, try to build loads of stuff and it gets really messy like this.

Whereas if you just have focus, you might have one project, but you get there a lot faster because you can just go from point A to point B with less friction, less complexity and everything's easier. So I think that's definitely the right move. This is an interesting one. So if you're not familiar with Pinocchio, basically it can install stuff for you, right?

I've actually not used it since Cloud Code came out, but it can install Hermes for you with Pinocchio, right? If you're not familiar with this, it's basically like an app installer for stuff that you would normally install via the command line or via terminal, right? And so you can install Pinocchio, it's free, and then you can set up apps directly. So if anyone's watching this and they're struggling with Hermes, this might be a really good option.

This is a really good question. So Vinay asks, what's the biggest AI automation that saves you the most time consistently? And what's working for everyone else? So for me personally, I would say setting up the AI avatar system was really powerful because it saves so much time.

And also if you imagine like each piece of content might take three or four hours to prepare, plan, edit, implement, et cetera. Whereas number one, I don't need to be involved in that anymore. And then number two, I can get my team to handle it too. So it saves so much time.

And we've actually got the full training on it here. Yeah, so we have the full training video set up on it right here, as you can see. So with many, I know you asked about that. We've got it right here.

Connor was asking about, is there an open core training course? So we actually have a full six hour course as you can see here, and also a full two hour course on Hermes as well. Lauren says, Omi looks amazing. Just saw this, it's mind blowing.

Anyone doing this, how does it compare to Seedance? So Enrique says very good, but not Seedance level completely, right? So if you want to get access to the new video model for Gemini with Gemini Omni or however you want to pronounce it, you can just go inside the chat here, click on create video, and then you'll see create with Omni, and that's how you can get access to it. It looks really cool.

I wouldn't say it's like Seedance level because from what I've seen from Seedance, it's absolutely wild, but it's really good. It's supposed to be like right up there with something like a Nano Banana and still a big step up from Veio 3. This is a pretty interesting one. So Damon was saying, is there an auto responder for school?

So like for answering questions inside the community, I've come up with two interesting ways to do this. Number one is that you can actually use Claude with the browser Chrome extension, and then it can actually go off and edit stuff for you. So for example, you can actually control your browser with Claude, and then Claude can post on school for you, right? That's one option.

Method number two is that you can actually get a custom GPT setup, and that can answer questions for you. So this is a bit more manual, but you can easily get a virtual assistant to do it for you. And basically they would paste in the question from the direct message, paste it into a custom GPT, the custom GPT is trained to handle it for you, and then that answers back. There's no API for schools.

I don't think you can automate it directly, but that's the best way that I've seen. And Damon says, where should I market my new AI SaaS? What I would recommend here is you wanna do like videos daily showing how to use a tool or using, for example, relevant tutorials where you show how to use your tool and what it can be used for. And then people will see that and they'll become interested in your tool and you'll generate more leads that way.

Also, if you repost those videos, if you post to social media about those sort of topics as well, and then for example, like maybe you do some blog posts on your website too, that would help you a lot too. But the main thing here is just being consistent with getting attention about your tool, showing how to use it, showing some cool use cases, and then directing them to your website. So here's an interesting one. So Karen says, I'm hitting a wall with my Hermes agent setup, right?

Basically, she's been like meticulously training there, setting up scheduled tasks, et cetera. And then she today decided to set up another agent for a specialized task, right? Created the profile, went to the workspace directly. And the goal was for it to interview her and get specialized context, right?

Now, after this, it started playing up. So here's what I would do. I would roll back the changes that you made, get rid of the new agent that you created. And also, if you can simplify your systems more, right?

So if you're giving it too much information, too much context to work with, it may struggle and it may get confused. So if you can simplify this process as much as you can, it's gonna help you a lot more. And then also decide, do you need an orchestrator agent or can you just have the agent directly doing all the tasks for you? So for me, for example, you'll see a lot of my workflows, like I rarely use a full orchestrator team of agents.

I just have one agent per profile, right? So for example, if we go to here, we've got one for OpenCLAW, one for Hermes, one for Cloud Code, one for pre-Cloud Code, Antigravity, et cetera, right? We rarely have profiles or multiple profiles for the whole Hermes. The only time I would do that is if I go to Kanban here and I'm just trying to create a team of agents.

Otherwise, you'll notice I always try and keep things simple. Otherwise, what's gonna happen is like your agents get confused and they struggle with everything else. So the goal is always to try and simplify as much as you can rather than giving it too much to work with and then it gets confused. And we have one final question here from Jason.

He says, how do you find the best setup for all this stuff? So if you're using Cloud Code or Gemini or whatever, if you don't have a lot of resources. So if you want to reduce the resources required, I'd recommend that you go with a free setup. For example, like you can plug in an API like Owl Alpha with OpenRouter.

And that's a free API you can plug into, for example, Cloud or for example, NA10. And you can see Owl Alpha right here. So that's how you can get a free model. What else we got here?

Surf says, is there a setup guide on how to set up Grok and Hermes agent? So we have a full training on that over here. And actually, it's pretty cool. Twitter's development team, they actually released a step-by-step guide as well.

So that's basically it. I've answered all the questions. We've been through everything. I've shown you how to use Grok with Hermes, the best free setups, the best automations to use, et cetera, how to use Hermes agent.

If you want your questions answered in video tutorial like this, feel free to post inside the AI Profit Boardroom community and answer questions like this. You can see we've answered loads of questions today. You also get access to all of my best trainings on this stuff. So you can see, for example, here, we have a beginner to expert automation masterclass.

We have all of these new daily updates and new video guides and trainings like you can see. You also get my agent operating system inside there. So you can learn, for example, how to manage all of your agents in one place, how to build these amazing tools, this full mission command center that just makes everything 10 times easier to manage. And also you can claim me personally.

Plus we do four weekly coaching calls. You can ask questions and get help on live calls. Additionally, inside the map, you can actually meet people in your local city who are interested in AI automation just like you are. So feel free to get that link in the comments description or go to the AIprofitboardroom.com.

Thanks for watching. I'll see you on the next one. Cheers, bye-bye.