Hermes AI DESTROYS OpenClaw!

Key Takeaways

Today, I'm going to run through my best automations for Hermes, which is a new free AI super agent. It's very similar to Open Claw, but I would actually say it's a lot better and a lot easier to use and it just works. The other difference with Hermes versus Open Claw is that it's faster, it's smarter, it actually works with the automations I give it and it just feels a lot better to use. So, this is Hermes agent. It is free to get right now. First [snorts] time you try it, you might not like it. That's exactly what happened to me. Since I started using it every day, it's so much cleaner, so much easier to set up. I'm going to show you some of my best automations for using this, which is basically like the AI systems I used to run my business with the Hermes agent and what I'm using and how the layers work and everything else. I'm going to show you like actual use cases for this, right? For example, if you have a look at my Hermes agent right here and we just run it inside Telegram. So, it is pretty simple and easy to set up. And also, it fixes a lot quicker and a lot easier than something like Open Claw. So, you see here for example, we actually generated this thumbnail directly with Hermes agent, right? And it designs thumbnails exactly how I want them for YouTube. So, it can generate pretty nice images and one of the biggest differences between Open Claw and Hermes agent is that Hermes agent is self-improving, like naturally self-improving. You do have Open Claw and you can train it to be self-improving, but even then it struggles. Even then it struggles with context quite often as well, it just breaks. I'll show you an example of where it breaks. So, you can see for example here, I started using it again this morning. I was like, "You working? You online?" It said it went full caps lock on me, right? And then I was like, "Are you still working?" like 10 minutes later and it just broke. And that is the story I think every single person who's using Open Claw is is struggling with, right? You can see that happening right here. It's like, "You working?" It's not responding. Okay, let's try again. Is it working? Yeah, it starts working again. Great. And then Open Claw just breaks on me again. And I've seen this same feedback across the whole internet. I've seen it across X. I've seen loads of people switching to Hermes agent. It only [snorts] launched in February. It's already got 15,000 GitHub stars. It just seems like a much smarter way to run your AI agents. It's free to use. You can use it with local LLMs as well. So, for example, you could use this with something like Ollama running locally as well. And you don't have to run it on a VPS, right? You can see here that Anne says, "And Hermes, you have to host on VPS, whereas Open Claw you host not." It's not true. So, I don't have a VPS, right? I'm not using Hermes with a VPS. This is running locally for me. And and also, one of the cool things about this is it actually runs with a background sort of automation. So, what that means essentially is like you don't need terminal running with Hermes agent to keep it running. Whereas, quite often when you shut down Open Claw inside your terminal, it just stops working like in the cloud. And even if it is running terminal, then it breaks again, if that makes sense. So, that's why I'm switching to Hermes more and more. This [snorts] is live, by the way. Like I'm doing it live right now. So, what I want to show you here is some of the ways I'm using it, some of the cool stuff and just give you inspiration cuz I think it's absolutely amazing what you can do with this. And also, I'll show you my system that I'm actually using to like document the process and to start like really improving at using AI agents and how I'm going to really take it seriously. I'll show you another example of what we're doing here. So, here's another example. So, I showed you the thumbnail process and I walked you through that a second ago and I will go more in depth on that in a second. But you can see here for example, we've actually created this website, bestaiagentcommunity.com, right? And it's basically it's fully designed and deployed with Hermes agent, right? The whole page was created with Hermes agent. Now, why would you want to do that? Why would you want to create a new website when you've already got for example, I've already got the AI Profit Blueprint.com. Why would I want to create another page? Because it's really powerful for AI SEO. And what I mean by that for example, right? And I'll show you an example. So, and this feels really old now. I think it was last year that I actually created this page with Bolt, right? Back in the day when everyone was using Bolt and love all that sort of thing, right? And it's an exact match domain, which means it's targeting this keyword and it's likely to rank on Google. It actually ranks number one on Google and then it sends traffic to our agency, right? So, when people click on the book a strategy call, it sends traffic to our SEO strategy session funnel. Great. The only thing is like that took a lot of work to create that page, right? It's a lot of work, it's time-consuming, it's hard to find the keywords, it's hard to create the page and then also you have to upload the HTML and deploy it right directly on a custom domain. Hermes can do all of that for you. And so, for example, what you can actually do, I actually asked it to create a guide for the YouTube video that I'm creating right now. And basically, you can see here I actually deployed it to a subdomain via Netlify, right? And so, it gives me a subdomain like this that I can easily create. And then also, go back to Hermes here, it actually gives me a domain management link, right? And what that means for example is like I can quick quickly create pages targeting a specific keyword, right? Like this one. And then it will give me a link to the domain management section here so that I can quickly add a custom domain and get this deployed like within two clicks, right? And all I do is I just give Hermes the keyword on exactly how to use it. It can automate websites for me. And if we go back to this enterprise SEO ROI calculator, right? That's the keyword that we're targeting. If you actually put that into Google for that particular keyword, you can see it's ranking number one above all these other tools right here, right? And so, it ranks number one on Google for that target keyword. And if you go into Ahrefs just to check how much traffic this is getting, it's getting like 600 search volume, right? And so, you can see my page ranking that we created with AI right here. And so, we can create and deploy automatically with Hermes using the schedule task feature. We can schedule Hermes to create a new page like this every single hour targeting a new keyword that generates more traffic to my website, right? And the goal is to generate more traffic to my community for the AI Profit Blueprint. And that's two different automations I've just shown you right there, right? Number one was the thumbnails like this, right? And the other one was creating the websites like this and then ranking them with AI SEO. So, let me talk you through a bit how this works. So, basically with Skill Boss, which is a skill I've installed directly to Hermes. Essentially, what we can do is we can number one, we can chat with any AI model, right? But that's pretty basic. I don't really care about that. But what we can do also is we can generate images and thumbnails, right? So, for example, I can create like thumbnails whenever I want for any video that I want just by sending a message inside Telegram. Saves me a lot of time. Saves me hiring a designer, right? And it's just an automatic feature right there. What we can also do is we can build and deploy websites. So, we can create apps, we can create landing pages, we can create static sites. It actually sends them automatically to Netlify with one command, right? So, the agent actually writes the HTML, the CSS and pushes it live. Here's an example of this working in practice, right? So, this is the page that we created. It's targeting this keyword, AI Agent Community. And basically, all I do is I go into Hermes, I give it a keyword and I say create a landing page around that and then it creates a page, writes it for me, designs it for me and deploys it for me as well, right? And so, it's super easy to create a lot of like websites and rank them on Google targeting the keywords that we want. And that's what this game is about. And so, two different use cases there. Number one is the thumbnails and number two is the websites. Now, let me show you some other stuff that we can do with this. So, we can also monitor and find trending topics in our industry. It's pretty simple and easy to do that, right? So, the workflow and how this works is like every four hours, it's going to monitor my competitors' pages and give me the ideas to my Telegram, right? So, it looks at it, what pages of my competitors do we have? And then it gives me some ideas on, "Okay, we could target keywords around this or that, etc." And then it has a daily workflow here. So, for example, hourly, it's finding fresh opportunities for me. Every four hours, it's doing the competitor research and just finding actual content ideas for me. Every six hours, it's coming up with new AI SEO keywords for websites. And on demand, it can generate like thumbnails, landing pages, tweets, content, that sort of thing as well. This is all going on inside Hermes automatically. Abdul says, "Yo, good to see it." So, this basically how it works. Now, it's basically a flywheel because we can find trending topics, we can create content instantly, right? For example, like tweets, landing pages, content. It even gives me for example, like a breakdown of different Reddit posts I could create. I'll show you another example right here, right? So, it actually scans on the internet exact match keywords that I could create a new website around, right? So, it actually gives me the intent of that keyword, the domain that I could get to launch a website with Hermes, the volume, so how many people are actually searching for it, and whether it fits with the AI Profit Blueprint, right? Because the goal here is like to generate traffic to the AI Profit Blueprint. And so, it finds whether it's a fit as well. And then also, it gives me a hook on in terms of what we can talk about as well. So, that's another example. And then it also gives me recommendations. It says, "Okay, my final recommendation is this and this. They're the strongest picks. They have the highest search volume. Let's go with those." And so, this whole stack of using Hermes is mind-blowingly easy to set up. Sulav says, "I hope he was live, but it's AI." It's not the AI, it's me, mate. And then it's two different people who said this is yeah, AI but it's not it's me. So, as we keep going through this as well, some of the cool stuff that it can do here. So, it can also if I give it feedback, it's got all these different tools that it can use, right? So, for example here, it can browse the web, it can navigate the web, it can look at different web pages with the vision tool. And then it can use my terminal locally as well, right? So, it controls your terminal too, which is pretty awesome. So, it connects locally as well. What else we got here that works really well? This is pretty cool as well, right? So, what it's doing for me here is it's finding opportunities, right? So, it gives me a scan for the AR proper boarding different opportunities that we could create, for example, like a video about. Or, for example here, it will actually draft Reddit posts as well. So, if you click on this, for example, right? And we open that up in a new tab, it's actually created a post for Reddit that we could just like paste directly into Reddit or into Twitter and that sort of thing, right? So, you can see it right here. And so, it scans all the information that we could create content around, and then actually writes the posts themselves. We can post directly, right? So, it saves a lot of time. Really powerful, really easy to use as well. And the way that it talks as well is pretty cool. So, for example here, you can see it's I got the full style down. This is pretty cool. Here's my plan. Here's what we're going to do. Here's the first batch of keywords that we're going to target. And yes, I can deploy to Netlify as well, right? And so, you see how it's like quite human in the way that it responds. Whereas with Open Claw, I've just noticed it's getting more and more buggy recently. It's becoming less and less useful, if that makes sense. The other thing that I really like about this is it can it gets significantly better in just a few improvements, right? In a few pieces of feedback. And I'll show you what I mean in a second. If we come back to image generation, let's say for example, you're trying to create thumbnails and that sort of thing, right? Here's an example. So, inside Hermes, I said, "Okay, here's my next thumbnail that I want to create content around." And it creates this, but it's like totally the wrong lands It's totally the wrong dimensions for a YouTube thumbnail, right? The image is totally wrong. And so, what we did here is we give it a bit of feedback, right? It says, "For the thumbnails, they should be landscape mode and use the actual logo, blah blah blah." And then it generated this, right? And this was better, but it was not that good, right? It needs to be a lot better. And so, what we actually did here is we gave it more feedback in terms of the style, and I gave it some images to use, right here. And then it actually upgraded the thumbnail skill, like you can see here, and then created this, which is way better than it previously was. And then it created this, right? When I I said, "Start creating text." Now, that was a lot of feedback, but the cool thing is because Hermes is self-improving, because it's always getting better all the time, if we actually go to the new thumbnails it generates now, it generates them perfectly first time around, right? Because it's self-improving, it's always getting better and better. So, the point here is like it it not only is less buggy, it's not only very easy to use, it's not also very useful, right? But also, it just learns. And it learns at a rate that I've never seen any other AI agent do before. So, some really cool stuff in terms of what you can do with it, how it works, etc. I want to talk you through some other stuff that we do with this. When you connect it to Skill Boss, Skill Boss is a paid plugin, but when you connect it to Skill Boss, it can do other stuff, too. So, it can, for example, like create videos, it has access to web research as well, and then it can create like presentations and it can send emails, too, right? Using something like this. And it's got access to all the latest image generation models, AI video generation models, and chat models, too, which is pretty cool. >> [snorts] >> And then the final thing I was going to say was the automation layer, right? So, this is what runs automatically. So, if you look down here, we break it down into like different hourly, every 4 hours, every 6 hours, and on approval. What we're doing every hour is we're looking for new trending AI topics, viral moments, and traffic opportunities, right? And then it can actually build things like tweets, landing pages, and email copy. Now, when it does that, it automatically creates the landing pages, automatically creates the tweets, etc. So, I don't need to do any extra work or approve it if I don't want to, it just goes off and actually does it, right? Now, also, what you've got over here is that you've got the competitor meta monitor, right? So, it can watch your social media competitors, see what's performing above average. So, let's say for example, one of your competitors posts a banging tweet, right? With that tweet, it can look at your competitors and be like, "Okay, that's something you need to create immediately, like right now." And then the final thing is keyword ideas, right? So, it finds trending AI keywords with building exam match domains around, delivers the keyword, the search intent, the volume estimate, and the headline hook. All I have to do is approve it once it's done. And so, these automations are just running automatically like all the time, which is pretty cool. >> [snorts] >> On the landing pages that it creates, it's like creating SEO-optimized the meta tags, the schema markup, it embeds the VSL, it's got six CTAs on the page, it writes it in the style of like direct response as social proof as well. And it makes it mobile responsive as well. So, if we go back to this page right here, and we open it up in mobile, you can see that it's all optimized as well for mobile, too. So, you literally just give it a keyword, and then it deploys it to a subdomain, and we're good to go on this, right? Now, this is quite a good question, which is how expensive does it get in terms of when you're using it? And I would say it depends on the API that you use. Now, for me personally, like I was rinsing it this morning. I was setting up all these automations cuz I just got to the point where I was like, "I can't use Open Claw anymore. It's just it's breaking on me all the time." And I see everyone else complaining about the same thing. What I actually did here is I gave it a Open Router key, like you can see, right? API key for Open Router. And this morning it used like $7. And if we actually have a look at the activity in terms of why that is, if we have a look here, for example, it's it used GLM 5 Turbo a lot. Now, bear in mind, we were doing the foundations of this, so we were, for example, like setting up the automations for the first time, prompting it for the first time, like really giving it in-depth automations, building all these crazy landing pages and automations. It's not going to be like that all the time. It was just for a quick run. And also, if you are like if you want to be careful with this sort of stuff, and what would I recommend? You could use something like What you could do is you could have two APIs, right? So, you have the brain, which could be like GLM 5 Turbo, or it could be, for example, something like MiniMax, it could be Claude, if if you're okay with the cost on Claude. And then what you can do is have a sub-agent and just run that sub-agent locally. So, for example, you could link this to Ollama and just run Ollama locally with the sub-agent, right? And that way your costs are very minimal because Ollama is free, right? If you're running a model locally, and then you're using Claude kind of like the brain of the whole operation to to assign tasks to the sub-agents. So, that's the way that I would look at it. And then also, you can use stuff like MiniMax, it's pretty affordable, too, so that sort of thing. It's pretty cool, yeah. And then one thing I was going to say is sometimes I just want to be 100 transpa- 100% transparent with you, right? Sometimes it doesn't work. And I'll give you an example of what happened here, and this was a really weird one. I've never seen this before with any model. But basically, you can see here, for example, I assigned it a task, and it started doing the task inside Hermes, and then it was like, "Yeah, let me build that properly." And then 5 minutes later, it just not done anything. I was like, "What's going on?" And so, what actually happened here is the model itself, the API itself failed, right? You can see it right here. And that's a model error, that's not a Hermes error. And so, what's happened here is like basically the model itself, which was GLM 5, just totally broke. That's basically the difference. And yeah, you just need to be careful with that. Now, if you find that same problem, if you have anything like this, here's how I fix it. There's two ways I fix it. Number one, I'll just go into terminal and I'll open Hermes up inside terminal, and I'll say, "Here's the errors I'm getting inside Telegram, fix these for me." Right? Because Hermes is like self-improving, and it understands itself well enough to fix itself. Usually, that will work. If it's like a total nightmare to fix, or if it's just, you know what, I need something that's a lot smarter, then I'll use something like Claude Code to help me. But nine times out of 10, you can just get Hermes to to fix itself. And then it seems like when you run Hermes inside the terminal locally, it's running almost like a different instance of Hermes. So, for example, when you're using Telegram, it's not the same conversation history as when you're using Hermes inside the terminal, if that makes sense. Like they seem to be two different instances, which is good because then the Hermes inside the terminal can fix itself inside Telegram. Let's see what we got on the questions. So, what are the differences between Open Claw and Hermes? One of the biggest differences is that with Hermes, it's self-improving, right? So, if you look at, for example, like Open Claw, you can train it, you can create like a learning MD file. And with the learning MD file, it's decent at remembering stuff, but it still struggles. Also, Open Claw breaks, and we've all been there, right? You update Open Claw, and then you spend 2 hours trying to fix it yourself. So, that's one of the reasons I stopped using Open Claw. >> [snorts] >> With Open Claw, it seems to get really messy now. You can have five different instances if you're not careful, especially if you're getting Claude Code to fix it with you, which means like when you're using Open Claw, it gets quite messy. One of the other things that's super annoying here is like, for example, if I open up the UI, the terminal UI, sorry, the gateway UI for Open Claw, you can see that we have to put in a gateway token to log in. And if we don't have that, then it fails to log in. And so I can't just access it straight away, if that makes sense. Whereas, for example, if I'm using Telegram with Hermes, it just works straight away. Now, you might say, "Okay, but why didn't you use Open Claw with Hermes?" Sorry, Open Claw with Telegram. But you can see here that Open Claw with Telegram just breaks on me all the time, like three times in the last week. So, that's why I don't use Telegram with Open Claw. And I was using before, but now you have to keep adding the gateway token and it gets super messy and annoying and it slows me down a lot. And then also the biggest difference is like the architecture. So, Hermes is designed to be self-improving, right? It's designed to create its own skills from experience. What that means is like every time it makes a mistake, it's either going to update the existing skill or it's going to set up a new skill. And then it seems to have a better memory. When you're using Hermes, it seems to use much better memory. The one thing I'll say with Open Claw though is like the community is far better. It's not I don't know I actually messaged my friends this morning cuz I was like trying to figure out, okay, who's Is anyone else switching to Hermes or is it just me? And I see a lot of people on Twitter switching, but I don't see many of my friends switching. I think that's because like usually I'm one of the early adopters. But when you look at, for example, Open Claw, it's got 340,000 stars on GitHub. When you look at Hermes, it's only got 15,000, right? So, the community support is not going to be as good. There's not going to be as much going on. This It's going to be harder to get in there and and to get all the support you need. Having said that, what I need is something that actually just works and it doesn't matter how many people are using it in the community if it doesn't work when I need it to. And so that's one of the reasons I'm switching to. What else would I say with it? No, I I think it just gets better. And you might be wondering, okay, like how does this work step-by-step? The other thing that I would say it's got it's got a terminal UI, right? So, when you're using, for example, Open Claw, you don't really use the terminal UI, you use the gateway UI, but the gateway UI is harder to access now. And you have to run a new terminal command to get the access token. And then with the access token, you have to paste that into Open Claw and it just gets super messy and time-consuming. When you use Hermes, like you you can literally just go into terminal, type in Hermes, hit enter, and it runs the terminal UI for that, right? Also, it works with all your main messaging apps. So, I think this is basically the same as Open Claw. You might, for example, get What would I say here? You might get more messaging platforms with Open Claw, but honestly, there's only one that I need or two, Discord and Telegram. They're the two that I want. Also, it's got a closed loop for learning. So, if you actually read that, unless you're super technical, you're not really going to understand it. But basically, it just it gets better the more you use it and it always is going to update the skill, I'm going to improve every time. You saw that with the thumbnail generation and that sort of thing. And then also What else we got here? Yeah, it's got a schedule automations, but that's the same as Open Claw. You can spawn sub-agents as well for parallel work streams. I've not really tested that much, but like I was saying, you could use Llama for that as well. And then you can run this with six terminal commands as well, which is pretty cool. And yeah, and I think also it's designed for research, too, which I've not really tested, but that sort of thing. The other thing I would say here is like when you're using Hermes, it doesn't seem I haven't seen anyone complain about the security problems with Hermes. Whereas, for example, with Open Claw, there was a lot of that stuff, especially when it first started. So, yeah. That That's some of the biggest differences. The final thing that I would say is this is run by News Research, right? So, News Research are a big company that have a lot of investment. They're like a big This is their biggest project ever, but basically with News Research, they're backed by some really big research, some really big partners in there. And they have a They're like a really well-respected research lab, right? So, if you go to News Research on GitHub, you'll see like all the other stuff that they've built here. And they've created a lot of different stuff here. Yeah, this is pretty cool as well. There's actually a paperclip adapter. I've not even tested that yet, but you could link this to Paperclip directly, which would be pretty cool, too. >> [snorts] >> You might be wondering, okay, what is Paperclip? It's basically like a way to run like a an AI agent company. So, you have all these different AI agents working within a company and there's actually like a paperclip adapter for Hermes Agent as well, which is pretty cool. Ah, there we go. Axis asked, right? Have you found any similar UI for Hermes, e.g., Paperclip for Open Claw? Yeah, so you've got this adapter right here, right? So, if we actually wanted to set up Paperclip with Hermes Agent, we could do that. So, we can just take the GitHub here. We can go back into Hermes Agent like that. We can say, "Okay, use this to set up Paperclip with Hermes." And that should integrate pretty nicely right there. Axis says, "Nice. We'll have to share that otherwise. Tried to merge it." Yeah, that's it. And then what am I using for the dashboard? I don't I don't I'm not using the CLI mostly. I just use it inside Telegram. I prefer the setup with Telegram rather than using the terminal UI. But I think a lot of developers would like the terminal UI. Like it's pretty nice, pretty easy to use and that sort of thing. So, you can see here now it's beginning to set up Paperclip. I'd be interested to know how long that takes. So, we set it up at 10:28 and it's already running the terminal commands to get it fixed. So, that'd be pretty cool. Axis says, "Yep, currently running on web UI. CLI is working, but the Ah, that's interesting. What are you using for the web API?" I'd be interested to know what you're using for that. Cuz I think when you're running through the onboarding with Hermes, it it asks you if you want to set up, for example, Firecrawl and all these different web search APIs. So, it's beginning to to set these up directly here. Now, if you're wondering like in the meantime, if you're wondering, okay, what is Paperclip? Let's have a look here. So, Paperclip is basically like a AI agent mission control dashboard, right? So, you've got your whole like team, but I'm not only that, you have these agents and one of them can be like the CEO, for example. One could be the CMO, right? So, for example, when I was talking previously about like how I've got an AI agent that automatically creates landing pages for me each day, I could have an SEO agent team directly inside Hermes that's constantly creating content, right? Constantly pushing new stuff out and that sort of thing. So, that's how I would look at it. Yeah. Now, some people say, "Why don't you use like Mana or something like that, for example?" And the difference for me is like when you're using something like Mana or these other automations, number one, like they tend to be a bit more restricted in terms of what they can actually do. Like they tend to be a lot more limited. And then also, when you're using, for example, like Max Claw or Kim Claw and that sort of thing, you just don't have full control over it, right? So, for example, if you end your subscription with them, then you don't keep it. Whereas, for example, if you're using Hermes and you're running that locally, all your files are stored locally, all your memory is stored locally and that sort of thing. Also, in the meantime, I want to show you what I'm doing with my journal here. I think this is going to be fun to have a look at. And I want to show you like, okay, here's what I'm doing, here's how I'm working and that sort of thing. So, I've created this little journal here, right? And basically, inside this journal, each day, what I'm going to track is like the progress I make, how I'm improving, what I've done, and basically, what's my to-do list. So, you can see, for example, each day, I'll copy it like this, right? And then I'll put the date. So, for example, like 29th March, I've already done today. And then I'll put the day number. I just want to track my progress in terms of, okay, what If this is day and I'm already on I'm only on like day two with this stuff, but if this is day two, what can I build, for example, on day four or day 60 or day 90, right? If I track all my progress, if I keep a to-do list here, if I keep a track and a journal of all the skills that I set up as well, which I've done there, I think this is going to become like really insane. You can build some amazing stuff. Like already just in a couple of days, I've set up so many cool automations with this that are actually super useful. I'm really excited to see, okay, what can we do in the future? What else can we build, etc., right? That's going to be fun. Now, if we go back into, for example, Open Router here, you can see what APIs I've been testing with this. So, I think we use Minimax a little bit here, too. Minimax M2.7. I want to see when we use that. What time did we use that? Yeah. And then we've got Claude Sonic 4.6. And GLM Turbo I've been using a lot as well, just cuz it's faster to respond and that sort of thing. Now, if we actually go to my dashboard inside Paperclip, you can see an example right here, right? So, for example, this is what it looks like. And basically, we set the goals. So, we could You can create a company with your AI agents. Some We're going really deep here. >> [laughter] >> This is I didn't plan to even cover this in the video, but you can see here, for example, like we've got a CEO, we've got a CMO, and they're just running 24/7, right? They're just going off, doing all sorts of stuff, building all sorts of stuff. And you can configure them so that you tell them, okay, here's your title, here's who you report to, here's your capabilities. Here's the AI that you're actually using. You can switch to Hermes as well, if you want to. And then this is running like 24/7 with a team of AI agents, right? So, you've got, for example, like a Open Claw, the founding engineer, CMO, COO. And then you've also got like a organization chart here, right? So, as you build more and more agents, you create like a bigger and better company where they're all managing each other. You can also control your skills, so what skills you have set up, the cost so far, the activity and what's been done. And then you can also change like the settings on your company as well. The most important thing I think is the goals, right? Cuz if you for example give them a really good goal and they work towards that, they're going to be way more focused. And these are all AI agents working directly with us, right? And so that's how you can set up paperclip. You can see here in inside Hermes it's taking quite a while to set this up. So I think that might take like maybe 15-20 minutes. That seems to be the process it's taken. But yeah, that's basically it. I've shown you how to set up Hermes, shown you why I use it, shown you all the cool stuff you can do with it, shown you all the automations I've set up, how to link it and create your own company for AI agents using this too with paperclip. And then if we go back to our dashboard here, you can see all the details of what's going on, right? >> [snorts] >> And it's completing tasks, it's getting stuff done, it's building stuff out, etc. So completed task right here. And then you can see like how many agents you got. But yeah, this just there's so much you can do with this. It goes really deep. I would just look at okay, what is actually useful for you day-to-day and then how can you automate that, right? Ah, this is really good as well. This is a great question. So Axis says, "Did you migrate any skills or tools?" Yeah, so it the skill boss one was the one that I migrated. But what I've actually done now is I've started creating a backup of all the skills I create cuz sometimes if you're fixing up for example if you're fixing open-claw sometimes especially if you're doing that with Claude code, like it just archives all your files or it renames the file names and then open-claw loses track of them and then you lose track of all your skills. So now what I do is like I keep a track of the skills that we're installing and setting up inside these agents. And then from there I make sure that >> [clears throat] >> I I basically save them to a document so I can import them at any time because bear in mind that Hermes just came out last month. Open-claw came out in January, right? It's like there are new agents coming out all the time and there's probably going to be something bigger and better than both Hermes and also open-claw in the future. And so I think it's really good to just keep a backup of this stuff and keep a backup of what you're working on and how you're using it and that sort of thing, So that's basically what I do. Here's an example, right? So you can see inside this document if I quickly whatever AI agent I switch to in the future I've got the thumbnail skill here. It's here's what I do. Here's how to create the thumbnails. Here's a screenshot of what I want. And then it can just quickly create those whenever I want, right? And I can plug that into any system that I use as well in the future. But if you do want to migrate, there is a migration option, right? This is pretty cool as well. It's there's an option for migrating from open-claw. So if you're coming from open-claw, Hermes can automatically import your settings, your memories, your skills and everything else, right? So during the setup wizard it detects open-claw and offers to migrate automatically. So it will actually look during the setup wizard if you're doing anything from open-claw and then everything that you're doing on open-claw is good to go. Arvin says, "Hermes is better than open-claw." That's it. I 100% agree with you, man. Thanks for agreeing. I'm glad it's not just me who's thinking that. If anyone else is thinking like okay, is uh Hermes is obviously better than open-claw. I'd be interested to know in the comments like whether you think it's better or not, but for me personally I do. I do think it's a lot better. Uh I says, "Reports on things breaking on the migrate." Feel I'd ask, but yes, keeping a backup makes sense. Yeah, that's it. And that's what I think is like you even on the migration you got to be careful, right? Now if you want to migrate it after you've installed this, let's say for example someone is watching this and they want to migrate, but they didn't do it during the setup. You don't have to run through the whole setup again. So you can use Hermes claw migrate inside your terminal and that will actually run through a whole migration process with you. So you can do that instead if you want. And then it will import like the sole MD, memories, skills, command allow list and everything else, which is pretty cool. So that's it. That is my whole guide to Hermes, what I've learned. I'll keep you posted in the future if people like this stuff. If you want to learn more about this, we actually got tons of training on Hermes and open-claw and Claude inside the AI Profit Room, right? So if you want to learn how to use AI agents to grow your business, to save time, to keep scaling and you want to see all the best stuff that I'm doing with this, plus you want to be able to directly connect with me, you can do that inside the AI Profit Room. It's got an awesome community of 2,700 members. We just hit 2,700. That's amazing. So you can post questions inside the community and get help and support whenever you need to. You can also join our daily accountability groups and each week I do a rundown. You can imagine what's going to be on the rundown tomorrow in terms of Hermes so that we show you okay, here's what is working, here's what you can ignore and that sort of thing. Uh I think you'll love that. If you're into AI but you just don't have time check out weekly updates I do every single day. You can actually filter down here to weekly updates and you'll see all the updates I do every week. So that will save time. >> [snorts] >> And then inside the classroom you get all of my best trainings. The main thing is if you check out the SOP update section here you'll see all my new trainings on all this new stuff that's coming out, right? So for example we got like a full 6-hour course on open-claw, another 3-hour course right here. And we just show you everything that's working. So for example I actually run you through how to set up open-claw, what the differences are, how to set up scheduled tasks. I've got a 30-day roadmap right here for Hermes which you can use. It's called the perpetual agent plan and you can use that with the day-by-day prompts to just use Hermes inside your business like I am to save time. Honestly, it's so much better. I think yeah. Axis says, "Any plans to make a Discord or are you only keeping it on school?" Just keeping it on school just cuz I think it's that much better place to keep things organized. Where do you think Manus fits compared to open-claw and Hermes? Like I mentioned before, like if you're using open Manus, you don't control everything and also like as soon as you end your descript your subscription you you don't have access to the agent that you spent maybe 6 or 12 months building. The whole point is I think these agents compound on each other massively and the knowledge and the skill sets you have compound on each other, too. So when you're using this stuff like you want to have something that you can always come back to, right? Like you wouldn't put your journals anywhere else apart from with you. And I think it's the same with agents. It's like you want to keep all that stuff to you because the more you use it and the more control you have over it, the better those agents are going to be. And also then you get more flexibility over which models you use, how you use it. If there's something amazing that comes out of open-claw Manus doesn't have it, but all your data is inside Manus, well it's like how am I going to switch? How am I going to set up those automations? It's going to be super messy, right? There's no migration option. So that's why I think open-claw and Hermes are better and specifically Hermes, honestly. I'm over open-claw. It just it was amazing when it first came out and it seemed a lot easier to set up when it first came out, but now it just seems like it's really difficult and technical to use. I have a lot of respect for Piers Stanberger and I think like he must be under a lot of stress as well cuz so many people are using it. So I don't blame him. I just think it's one of those things where they added a lot more functionality to it and in the process the magic of open-claw got lost a little bit. The Harish says, "I missed it. Can you start over?" You can just rewind. You can rewind on the stream. Abdul says, "Great live. Thank you very much, sir." Yeah, that's basically it. And then also if you want to get more clients you can check out the agency course. If you want to get like weekly coaching courses stuff like this just join the live calls inside the calendar, right? So you can join all these live calls and get help and support real time on those coaching calls. And then also if you want to learn how to rank number one with AI SEO, you can check that out here. If you want to learn how to grow your YouTube channel, you can check that out, too, right? So that's all inside the AI Profit Room. Final thing I was going to say was like if you want to connect with people in your local city, your local area, etc. Because we have so many members around the world, pretty much every major city we've got people in there, which means like you can see them on the map, you can connect with them, DM them, meet them in real life, etc. It's just a great place to to bring people in. Thanks you work you on my face, my face. Very much, sir. All right, appreciate everyone who joined today. That was an awesome live. I'm absolutely loving Hermes. First time I've been excited about AI this much since open-claw first came out, right? And open-claw was amazing. Now it's got worse. Hermes is better. Let's all use Hermes. All right, see you on the next one. Cheers. Bye-bye.