Transcription
In today's video, I'm going to be walking you through step by step how to set up Hermes agent, which is the 24/7 self-improving AI agent that really has been taking the place of tools like Clawed Code and Open Claw. I'm going to be showing you how to set it up so it's always learning from itself 24/7 and just give you a basic rundown on what it is. That way, you can understand if it's something that you should be implementing in your business right now or not.
Okay, let's talk about Hermes agent. The self-improving 24/7 always on AI agent. What is it first of all? Well, it's an autonomous agent that lives on a server. Doesn't just have to be your server or your machine, but it can be a cloud server so that it can stay on 24/7. And this remembers what it learns. This agent remembers the feedback you give it. And it gets more capable the longer it runs. So, it's not something that you set up today and it's instantly useful. In 30 days, that's when it starts to become useful. In 15 days, it becomes more capable. The more feedback you give it, the more it learns. And so that's Hermes agent. It's not like Clott Chat GPT where it just remembers little things about you in its memory and then it operates. This is remembering how you get tasks done.
And so Hermes agent really has three layers, and I want to break it down in these layers so you can really understand the general architecture of what makes this so powerful before we get into deploying it. This is something super useful that I recommend everybody studies and learns before just jumping right into it.
So, layer number one is memory. This is one of the most important things to understand about Hermes agent is that it has a very good memory. When you correct it or it learns something stable about you or your setup or the way that you get things done, the way that you complete a certain task, what it does is it writes it to persistent memory immediately, which means it gets injected into every future session and every future conversation you have with it. So, if you're trying to complete something, it's going to constantly be learning about the way that you like getting things done that actually drives results, not just theoretical things or weird things about you, like what you like to eat, right? It's not what this is about. This is about how can it get you results? What does success look like? And how do you contribute your feedback to success? It captures all of that in a nice memory setup.
Layer number two is skills. And skills are basically automations or miniature workflows that AI agents like Claude Code, Codeex, Gemini, Grock, everybody uses this skills format. Now, Hermes has skills, and when Hermes solves a problem that it hasn't seen before, it writes down exactly how it did it so that next time that situation comes up, it loads that playbook instead of starting from scratch. You can read it, you can edit it, you can add your own. Skills are something that Hermes has integrated automatically. They've got a bunch of different skills that they're adding pretty much every week or almost every day now where they are adding these pre-built skills that you can use. So, they already have an automation library pretty much baked in that we can access that I'm going to show you a little bit later in this video when we set it up.
Now, the third layer is cron jobs. Now, cron jobs might seem like a code developer-styled term, but it's nothing to be scared of. It's basically tasks that run on a schedule. So if you set up a cron job, that's basically you telling Hermes or Hermes telling you that, hey, I'm going to run this task every day at 5:00 p.m. or I'm going to run this task every 5 minutes or I'm going to look for something to do every 5 minutes. And you basically describe your job once that you want completed. You set the time you want it completed at, and it will fire automatically on the Hermes server with Hermes agent. Whether that's every morning at 8 AM, every five minutes, every time a new video from a certain channel you like drops, this is your Hermes agent's trigger. It can constantly pull and look and turn on and check for certain things pretty much instantaneously or whenever you want it to. And the cool thing about Hermes is sometimes it will set up its own cron jobs, its own schedules depending on what it thinks is best for your situation and for the target that you're trying to accomplish. And so these are really the three layers that make up Hermes agent. We have memory, we have skills, and we have cron jobs.
The Hermes loop in action starts with a target. And you are always the one defining direction, defining the target. You can set up Hermes to define miniature targets, but the general goal of you getting into this, there should be a goal behind why you're using it. I don't recommend just using Hermes agent to try to build a million-dollar app in the background. You can try, and it'd be a fun experiment, but you might be wasting your time and money.
Now, after you set the target, everything stems from that. So, Hermes, when you set a target, you say, "I want to accomplish X." What Hermes is going to do is it's going to read its memory. It's going to read its skills, and it's going to start deciding on how to best help you based on everything it knows about you. And so you can see why this gets better over time because in the beginning, when you use Hermes agent, it's going to read memory about you, and it's not going to know anything about you. It has a nice skill library already baked in, and chances are it can help you out of the gate. But these aren't custom to you, and so you have to work with it for a little while in order to make this useful.
But after it reads its memory and skills, what it's going to do is it's going to use tools in order to try and accomplish the target you set. These are tools that you set up. It's tools that it has baked in. A lot of tools that are going to be useful to you are connections to your existing applications, things like Gmail, GitHub, etc. After it uses the tools to try to complete the action in order to accomplish the target, what you get as a result, and this result's important because this result is going to get better over time. So this might be a fixed target, but chances are this should be an ongoing target, something that you need an employee for, something that you need improvement upon, a new revenue stream that you're working on based on your existing domain expertise and experience. This result is going to try its best to get to that target.
And so what happens after the result is you provide feedback. And this can be you physically typing into your communication medium, whether that's Telegram, Slack, iMessage, WhatsApp, however you're communicating with this AI agent. You can type feedback based on the result that it got. You can say, "Hey, I didn't like this. I liked this. You did good. Here are the analytics." Or you can have it check analytics live. For example, if I had it creating thumbnails for me on a YouTube video, what would I check or what would I feed it so that it could see how good it did? Well, I'd probably feed it click-through rate, right? How many people are clicking on that video? If I have this Hermes agent creating thumbnails for me, is it a good click-through rate? Is it bad? That's stuff that Hermes agent needs to see. What's the industry average? Did it do good? Did it do bad?
And then the cool thing about Hermes is Hermes will update its memory and skills based on the feedback that you provide or the feedback that the analytics provide. I like thinking in terms of analytics because analytics are very black and white. It's either good or it's bad. And feedback is good for things like content, things like maybe you didn't like the way it scheduled a meeting for you. Maybe you didn't like how it committed a certain line of code to your code base. Maybe it did something wrong in security-wise, whatever it may be. Hermes will update your memory and skills based on the feedback. And then it injects it back into the memories and skills. And so this target is a never-ending loop. And Hermes will always see the target and it's always going to be updating based on your feedback. And once you get the loop good enough, it's going to run by itself to the point where you can almost skip feedback. That's always the goal: how can we skip feedback and go straight from result to starting the loop again.
And so this is really the Hermes loop in action and everything that you need to know in order to understand how to best use this and kind of the direction that you need to start thinking in when it comes to setting up Hermes agent.
Now, there are two main ways of setting up this agent. You can do it locally, or you can do it on a VPS, which stands for a virtual private server. You can think about it like you're connecting to a computer in the cloud. So, it's not going to be taking up the processes on your computer, and your computer doesn't have to be on. We're going to get into that, but let's compare it on a couple of different factors. Today, I'm going to be setting it up on a VPS. I'm going to be showing you because it's the most scalable framework for pretty much anyone to just jump in and start using this thing and having it work all the time. But, I want to kind of compare these two ways of setting it up so you can understand the upsides and the downsides of both.
So, when it comes to local, the cost of local is free because you can just set it up on your computer, and your computer is the one manning the storage, manning the costs, the RAM, all of the process would run on your computer. So it doesn't really cost anything other than the hardware that you already have. So that's very nice, especially if you are technical and you have a nice setup. If you want to run it on a VPS, that's going to cost you money because that has like energy and computation that it needs to be running. And the cost of that is $5 to $10 a month. Pretty trivial for what the Hermes agent can do.
Now, let's compare to always on. The problem with local is that it's not always on. The second you take a flight, the second you close your computer, the second you go to bed, if your computer's not on, your agent's not on. And so, that is the downside of local. But with a VPS, it is on 24/7 since it is in the cloud. You always have your agent learning. And that's what brings me to the self-learning topic right here. One of the biggest reasons you'd ever want to use Hermes agent in the first place is because it's self-learning and always on. And so those are two very important factors. Now, Hermes is self-learning on local when your computer's open, but on the VPS, it's always self-learning. It's on 24/7. So you're getting the benefits of this agent in its entirety.
Let's look at setup. Setup is both one-time. It's not like you're going to be bouncing back and forth. The VPS, I will say, is a little bit more complicated to set up, but that's why I'm going to walk you through it on this video, step by step, and I have a setup guide, one-click install, super easy stuff for you to get up and running with your Hermes agent.
Now, in terms of flexibility, I would give local a low amount of flexibility unless you are very technical or unless you have multiple computers that you don't use and you don't mind just having them set up in your basement running 24/7 or in your kitchen, wherever you do that. So, flexibility on local is low because you really only get the benefits of what your one computer can do, and it always has to be on. But with a VPS, you can set up multiple AI agents. Your computer doesn't have to be on. You don't have to worry about your power going out. You can just let it run all the time. Multiple different agents, multiple different processes, and it scales with you.
And so, let's get into setting up this Hermes agent from the ground up. Not only am I going to show you how to set it up, but I'm going to be showing you a live example at the end of this video of how I am using it in my coaching business in order to get traffic to my paid offer. And so, I think that's going to be very beneficial for you to see how someone who is currently running a business, current, who is in AI, is actually using this.
All right, let's get into setting up Hermes agent. I'm so excited. And honestly, this is the hardest hump to get over, but once you get over this, everything from here on out is pretty smooth. And the agent just improves when you use it. And so, it's very nice, very easy once you get it set up. So, let's learn how to do that.
We're going to be using Hostinger, which is today's video sponsor. And it's super convenient that Hostinger is a sponsor because this is something that I would use regardless. So, Hostinger is a company that gives you your VPS, and it's super nice because they're also in the AI space and they're always looking for things like Open Claw and Hermes Agent, for instance. So, they have easy one-click installs to make this process much more simple in getting your AI agent on a virtual private server. And so, I have this linked down in the description, and you can get to this page in order to deploy your agent. As you can see, it's as low as $8.99 per month.
Now, when you select a plan, I recommend selecting KVM2. That will be more than enough for your use case. You can set up multiple agents on this thing, and it offers a very fair price for what it can do. And so, when you have KVM2 selected, you can select deploy. And this is the beginning of setting up your Hermes agent. I'm going to hit deploy. I'm sorry for the light mode. I don't want that to blind you. But once you hit deploy, you are going to be brought to your cart. You can select the period that you want. Obviously, the longer you go, the more value you're going to get for your money. So, if you go 24 months, you can get this as low as $9 per month, and you're going to save $372. But I recommend just going with the 12-month plan. You basically get a year's access. And I can even show you how to get a little bit more money off because you can use my coupon code AI foundations, all capitals. I'll leave that in the description. And if you apply that, you're going to get an additional 10% off, which is going to lower the cost even more. You have other options here like a daily auto backup if you want to select that for an extra $6 per month, but I recommend you just get acclimated to the environment first and then you start worrying about things like backups once it's running half of your business for you. Then you can hit continue. And after you get done paying, you're going to get the congrats screen. Your journey begins now. And it truly will be a journey as you're using Hermes agent.
Next, your dashboard's going to load. You're going to get this Hermes agent configuration screen. You can just select deploy here. And then it will set up your VPS. And this is going to take about 5 minutes. And just like that, it got everything pretty much created for us. And in a matter of about 60 seconds, everything was deployed. Very quick, very easy. They made it super simple with that setup guide. So once it sets this up for you, it's very simple. If you go through that entire deploy process to start using Hermes agent, if you go to the Docker manager and you go under projects and you see Hermes agent in here, it means you've done this correctly, and it does it for you. So you should have done it correctly. But right here where it says access, what you can do is you can select terminal. And what this is going to do is open up a terminal for you on the web. And it's going to be super simple to get everything up and running because this isn't on our local machine. This is connecting to a terminal on our VPS.
And so what you want to do when you get here is simply type in Hermes setup. And then this is going to walk you through the setup wizard. And once you type that in, you can just hit enter. And since you're already in the container, it's now going to ask you how would you like to set up Hermes. I recommend just doing the quick setup. The full setup is a little bit confusing, and honestly, you don't really need it in order to get the most benefit out of Hermes from an early stage. So I'm just going to select enter for the quick setup.
And the next thing it's going to ask you is for an inference provider. So, what model do you want to use? This is one of the biggest benefits of using something like Hermes instead of Claude Code, for instance, is because you actually get no vendor lock. You can set this up with any model you want, including models like Claude. If you have an OpenAI Codex subscription or just an OpenAI subscription in general, what you can do is you can sign in through that subscription, and you don't even have to incur any cost, which is very nice that OpenAI actually does that. And so it will use your subscription for model inference.
But what I'm going to do, and what I always like to do, they've got a ton of models in here. It's hard to choose. And because of that, I use Open Router because Open Router is very easy to use, very simple, very straightforward. You get access to pretty much every single model on planet Earth, and you need one API key, and you can change them on the fly. And so I'm going to select Open Router, and then it's going to ask me for my Open Router API key. I recommend you do the same, unless you have a ChatGPT subscription, then I recommend you just go through the authentication process there. It's going to ask you to sign into your OpenAI account very simply.
And so this is Open Router. I'm signed into my account. I'm going to select get API key. It's super simple to get an API key. What you can do is just load in some credits. If I go to my credits tab after I sign into my account, you can just load in a little bit of credits. And you can load, you don't have to load in $100 or anything. You can load in $5 in order to get started, and you get access to pretty much every model on the planet. So, I'm going to go get my API key. Make sure you keep this a secret. I'm going to say Hermes 3. You can set a credit limit. I'll just set it to something like $25, and I will hit create. Obviously, you don't want to show this key to anyone. I'm going to rotate it after this video. I would not leave this out on the web because people will be able to use it. I'm going to hit copy. And now I have my key.
So, I'm going to go back to the setup wizard. And right here, I'm going to click in, and then I'm going to paste. And it won't show up because that's how secrets are in a terminal. It doesn't show up. So you can just hit enter, and it will be there. And then you can select your default model. If you want to spend as much money as you possibly can, then use something like Opus 4.7 or GPT 5.5 Pro, which is an insane amount of money. Honestly, I like using Kimmy K2.6 because it's pretty much like Claude, except it's like, I don't know, 10x cheaper. But you can use pretty much any model you want. I'm going to use Kimmy.
And then the next step in connecting to Hermes is setting up a messaging platform. Honestly, the easiest one is Telegram. Telegram is super easy to set up. And so we can just set up the messaging now. I'm going to hit enter. And we are going to use Telegram. So I'm going to type in that number to confirm. So I'm going to select one, and then I'm going to hit enter. As well as you can see, I mean, you've got a ton of different options here: Microsoft Teams, Google Chat, Matrix, Slack, Discord, WhatsApp, Signal, Email, SMS, you can do pretty much anything you want. Telegram's the easiest in my opinion. So, I'm going to set it up that way. And then I'm going to hit enter. It wants me to confirm. And so, since I've typed in one, I will just hit enter.
And now, we need a Telegram bot token. So, this is when you need to go to Telegram. Every app is going to be different that you want to communicate to Hermes through, but I'm going to go to Telegram since this is the one that I've decided to do for this video. And so in Telegram, what you want to do in order to get your bot token is you want to find the BotFather. And so you can go to apps and type in BotFather. And this BotFather guy will come up, and you can just select open, and it will take you to this screen. You can also do it in the chat as well, that's no problem, but this is just a simple way to create a bot. And so now that I've opened up the BotFather, what I want to do is select create a new bot. And then I can give my bot a name. I can name this whatever I want. I can just name this Drake PA for Drake Personal Assistant. Then I have to give it a username. So I'll name it Drake PA bot. Let's see if it's taken. It looks like it is available. So that's good. I will select create bot. And then it's going to give me an access token right here, which I can just copy very simply. And then right here where it says Telegram bot token, I'm going to paste that in. Again, it's not going to show. That's okay. You can select enter. And now it's saved.
Next, I need to add my user ID. And so this is my unique Telegram ID. This is basically going to gate access to this agent on Telegram from anybody else except me because I'm putting in my allowed user ID. So not anybody's just going to be able to message my Hermes from Telegram, but only people with my ID, which I am the only person with my ID. So, in order to find your Telegram ID, what you can do is you can X out of the BotFather and then you can go to a different app, and it's actually going to be a chat, and it's called User Info. What you can do is find this User Info Get ID IDBot and click into it. And when you type in /start, it's going to pop up with your ID right here. And so I'm going to copy that first line, and I am going to paste that in once again. And then I'm going to hit enter.
And now it's asking for home channel. It says, "Where Hermes delivers cron job results." And so this is the channel that Hermes is going to message on Telegram. And so I can just select Y, a capital Y for yes, and set my user ID to the home channel and hit enter. And now we are ready to go. It's going to say launch Hermes chat now. So I can hit Y and let's launch Hermes chat. You can just follow the instructions. I'll hit enter. And now it's going to load up Hermes, all ready to go for us. It's going to open up the Hermes CLI, which looks amazing. And just right out of the gate, you're going to get that dopamine hit because they give you the available tools. They've got all these fun tools in here that you can use that really no other AI agent can just do on the spot like Hermes can. What I recommend doing is just making sure it's online. Just saying something like, "Hey there, Hermes. What model are you using?" and entering and sending that off and just checking to see if your agent is online. And you can use it in this chat over here. It says, "I'm currently running as Moonshot Kimmy K2.6 via Open Router." And sometimes the CLI is a little buggy, so it looks like it duplicates messages, but I think that's just a UI thing.
Now, one thing you can do is you can type in /help. And this is going to show all available commands that you have access to on Hermes that you can use not only in this version on the web, but also on your messaging app like Telegram, which I'm going to show you how to get that all set up and going in a second. But I can hit /help, and there's just a ton of cool things you can do with literally typing in /something and having it do just crazy things for you. For example, /background. You can run prompts in the background. You can do /agents and you can show your active agents and your current running tasks. /Q, which will queue a prompt for the next turn. /goal, which is a super, super important command because this sets a standing goal for Hermes to work on across turns until it's achieved. And so goal is a big unlock. And remember, there are a bunch of skill commands. Look at some of these skills that you can use. You have 85 pre-installed skills.
Now, let's get into actually using Telegram with our AI agent because right now Kimmy's still listing all of its tools over here on the left. But I want to be talking to it on the go on my phone in Telegram. This is really where the real magic happens. There's a little bit of a caveat because just because you set up your gateway doesn't mean that it's going to be on 24/7 or even on at all. So, ensure that when you are within this Hermes agent chat on the web, which if it's glitching out and being weird and it's going up to the top of your screen like this and the messages aren't loading, the easiest way to get back to the chat in a normal setting is to go to your hosting panel and select this terminal button again and then just restart the process. Okay, if you've connected and you went through the startup process, now every time you come in here and you type in Hermes and hit enter, it's going to pop up with a brand new version of the CLI. And so if it ever bugs out, you can just do that. But the goal is to message it on your messaging platform.
And so in order for us to get it on Telegram, what we need to do is we need to type in /gateway and then we need to hit enter. When we do /gateway and we hit enter, a couple things are going to pop up. Number one, we've already set up Telegram. It's enabled. We have the homepage set up. We went through the process, but we need to start the gateway and we need to get it running 24/7. So, in order to do that, we need to copy this command right here. They give us a nice easy command to follow. And if it doesn't work, then what you can do is you can message Hermes, "Hey, why didn't this work?" This should work if it's giving it to you. But I'm just going to paste that in. Actually, no, you don't want to paste it in here. That's a mistake. You don't want to paste it into the Hermes chat. What you want to do is you want to hit Control C, and that's going to exit you out of Hermes and put you into the directory like a terminal directory. So now you're controlling the Docker container with code or with commands just as you would any other terminal. But what you can do is you can paste that Python command right here. When you hit Control C out of Hermes, you don't want to type this in in Hermes. It can do it for you, but it is faster if you do it this way. So now I can type this in. I can hit enter. And what that's going to do is it's going to start the Hermes gateway for the messaging platform. I've already done this, but what this is going to do is it's going to start that Hermes gateway for you and use that Telegram bot token if this is your first time doing this.
And so now if I go to my Telegram chat and use that exact same bot that I created, what I can do is I can hit start or I can type in a message and I can do whatever. And then it's going to be connected to the Hermes gateway. And so I can say something like, "Hey there, are you going to be connected 24/7 now?" And then I can send that off. And in the upper left-hand corner of this chat bar, you should see typing on Telegram if the agent is connected. So, as you can see, mine is connected and it is ready to go. As you can see, it said yes, it is on 24/7, but it depends on the infrastructure.
And so, now, let me just close out of this web chat over here on the left. And, you know, I could close out of everything now, actually, and just open up Telegram full screen. And now, I can have a full conversation with it. And so, I can type in something like, "I want you to become a personal assistant for me. I want to be able to create Google Calendar events with you and set reminders so you can notify me almost like a second brain X personal assistant. Here are some of my documents." And now I'm going to attach some files. And as you can see, I have a voice.md and a Drake POV.md. Basically, what this document is is a rule book for how AI should write when it's either writing emails, content, whatever it may be. For me, I've set this up so I can give it to any AI agent and drop it in. I basically told AI, I was using Claude, and I told it to interview me and talk about a voice and copy rule book and some guidelines that I have for writing copy with AI. And so I have things like a banned word list. I also have my POV document that I have all of my points of view in life because AI without a point of view, without any backbone, is just like everyone else. Everyone's using the same model. And so I have some of my hot takes, some of my contrarian takes, my actual beliefs. And I don't want to invent positions that I haven't stated. I just want it to speak who I am. And I really don't care at this point. I want this AI to be dedicated to me. So, we have my faith, we have my routines and success, we have AI and creativity, education. I even have a stories document of all my life stories, but it doesn't really need to know that. Sometimes it uses them too much. So, I'm going to hit open on those two, send it off, and it's going to begin training on me essentially.
And so, as you can see, Hermes, what it's doing is it's picking up on little things about me just from these two documents I uploaded. It took me an hour to create these documents with AI interviewing me. You can set it up to talk to you. You can even do it on Hermes. Tell it to interview you about your life and your point of views, and it's going to create memory for you. And basically, what it says is, "Here's what a second brain and personal assistant setup looks like. What I can do for you: Google Calendar reminders, second brain storing notes. What we need to set up first." And then it's just going to walk me through that setup process step by step and practically hold my hand through the entire thing. And once I start setting things up, having it create events, it becomes very simple.
Now, this could go a hundred different ways. You can do whatever you want with Hermes Agent. What I'm going to do is I just want to show you what I'm using it for live in production in order to drive real results to my business, AI Foundations, over the long haul. I'm not trying to get a quick win with this, as you should not try to do this either. What I recommend you do is you set up things that can run in the background that you know are high leverage that you just don't have the time to do that could eventually bring in a revenue source while training Hermes how to do a really good job at one thing.
So, I have my site here, and I have a blog that I'm launching basically for all of my YouTube videos. This is something super simple, and I told Hermes, "I want you to check my YouTube channel. Anytime I upload a new video, I want you to create a blog post around it. Upload it. Look at the analytics." I had to download this app called Plausible to my live site, but it's creating these blog posts for me based on my YouTube videos, based on my sole document, based on my POVs. And this is just something that's going to run in the background for I don't care how long, a year, two years, and it's going to be refining itself based on live feedback from how many people are actually coming to the site, clicking the links, and reading these blog posts, which is very neat. And it even links my YouTube video down here for a legitimate blog. Not just something that AI generated, but it's using my voice. It's using my transcript. It's linking my video. And so this is something that's hard for other people to replace, but it's also something I can set up in the background for a long time. And I can just let it run and let Hermes improve on itself from the analytics while I'm not there. Sure, I can revise some things, and I already have. As you can see, I even created myself a little admin dashboard, and Hermes kind of works out of here, and I can revise certain things. Maybe I don't really like this title. What I could do is I could select revise and I could say, "I don't like this style of title where it says stop doing X. It sounds too overrated and AI slop-like." And so I could revise the title, or I could go revise some of the body copy if I wanted. But then I can just select send to Hermes. And now it's in the revise section. And within 5 minutes, Hermes will have that back in the drafted section for my review once again. And this is the loop. If I like a blog that it generates, I can just hit approve, and then that will go in its feedback as well.
Just to show you that this is connected and kind of jump-starting the process, I can go down and actually talk to it, check my revisions, and make the change. This is actually pulling every 5 minutes. So regardless, in the next 5 minutes, it would pick it up. But I can also manually run it and talk to it that way. And now it's kind of asking for my feedback. It says, "This is the exact pattern you hate: imperative stop X, start Y. Before I write a replacement, a few options in your voice." And so it's going to keep brainstorming with me, and it would even message me and notify my phone. Since this is on Telegram, I'm getting Telegram notifications here. And now I'm kind of controlling my website and my content system all from what could be my phone. But right now, it's just a Telegram chat thread in specific, all because I have a Hermes agent on a VPS running 24/7 for me. It says, "Done. Title updated. Nothing else touched. Let's go check it out." And so if I go back here, as you can see, it's out of the drafts and it's now in the drafted section. I can click in here. I can review the post. Now I can just hit approve, and yes, approve. And so now that's accepted. I can even have Hermes look at the ones I accepted and learn from that as well. So it's always on. It's always learning. And it's because I have it hooked up to that VPS like I said on Hostinger.
So, this has been a full guide on how to set up your 24/7 self-learning Hermes agent, and it's only going to get better the more you use it. This was kind of a general video showing you the use cases, showing you how it works, showing you how I plan on using it, so you can decide for yourself whether you should use it or not, too. Now, again, if you want this thing running 24/7, I highly recommend getting it on that VPS. I have a coupon code for Hostinger down in the description, as well as links for that one-click install where you can just follow along and get everything set up. The best way to get working with this, like I said, is to get using it as soon as possible so it can begin learning everything about you. And the VPS is the easiest way to do it because I can literally talk to it on my phone, and I could be on a walk, just, you know, quickly tap a hands-free button and just talk about anything in order for this agent to get a memory about me and start learning about me. And you control everything, the skills, and all of it.
So, I hope this video was helpful to you. If it was, please drop a like and subscribe. I would highly appreciate it. And with that being said, I'll see you in the next.