Hermes + Aion UI is Insane (FREE)!

Today I'm going to show you how to run Hermes AI Agent inside Aion UI completely free and why this combo is one of the most powerful AI agent setups you can get right now for free. Because most people are using Hermes inside a terminal. It works, but it's messy, it's hard to manage, and it's not built for running multiple agents at once. Aion UI fixes all of that. It's a free open source interface where you can run Hermes, Claude Code, OpenClaw, Gemini, all of them in one clean dashboard. You can switch between them instantly, schedule tasks, control everything from your phone, even set up agent teams where multiple AI agents work together at the same time. In this video, I'm going to walk you through exactly how to set up, how to connect Hermes to it, and how to start using it to automate tasks in your business today. It's free, it works on Mac, Windows, and Linux, and once you see what it can do, you're not going to want to go back to terminal. Let's get into it. Today, I'm going to show you how to use Hermes with Aion UI. So we've got Hermes Agent over here, but we're going to be using it with Aion UI. Now, this is a free way to manage all of your AI agents. You can actually have, for example, OpenClaw, Hermes, Claude Code, even Gemini CLI and OpenCode working together using this process that I'll show you today. So the way that we can get started here is we just go to their GitHub for Aion UI, and then to install it in one click, I'll show you how to do that in a second. But you can see an example here. So it's like a nice little chat setup. You can switch between your agents at the top. You've got a nice UI. It looks a lot nicer than the terminal here. And it's much easier to manage everything as well, so you can control everything better. And you can still set up automated and scheduled tasks. Plus, you can link that to your phone as well, which is pretty nice. So to get started with this, you can either set it up on macOS like this, or you can just download it. That's the other beautiful thing about this is you can easily just get it downloaded. And to do that, we just scroll down to here, right? So you go to the latest releases. You're going to scroll down, and then you're going to download the relevant setup for you. It is available on Windows and Mac and Linux, which is great too. So I'm going to download it on Mac right now. And you can see that it was newly updated just five days ago. Once you've downloaded it, it's going to look like this. We're just going to open this up. And that's set up now. So if we open up A on UI, click on Open. This is a free and open source project, by the way, before you ask. And look at the UI here. It's just so much nicer than using it inside Terminal. I mean, compare these side by side, right? Here's your Hermes, which is great. It is an amazing agent. But here's Hermes inside the chat. Which one is nicer and easier to use? For sure, it's this one. You also get this Skills Marketplace. You can actually search for Skills over here as well, which is great. And you can attach it here. So what this means is that you can actually search through and find new skills for your agents and then just install them across multiple, right? So what it feels like, and I'll show you an example of what this feels like when you use it, is when you're using Claude Code, it feels like Claude Code or Claude Desktop. But the difference is this is free to use with A on UI. And Claude Desktop is amazing. But a lot of people, they can't and don't want to use it, right? So this is a better alternative. Now, also, we can add more models inside our settings. We can add assistance over here. We can also set up different integrations with different agents too. And if we scroll back to the main interface here, we can search for our conversations. We've got the Scheduled Task section. And we also have the ability to switch between agents. So if I go to Hermes here, and I'm like, right, are you working? Just to double check it is actually working here. Sometimes as well, you have to bear in mind, like the UI might be set up, but the integration between Hermes and A on Web UI is not set up sometimes. So I'm just going to double check that. That's why I want to check it's working right here. And we can also see that we have four loaded skills. So we can always create more later too. And then we can have multiple agents parallel over here. So we could have Claude Code over here. We could have OpenClaw over here, you see. And you can chat between them and see your history between them, which is awesome. Now, if we go to a new chat and we select Hermes 2, we can switch between the models that we want to use. So it will show your listed models that are set up in the dropdown. You can switch between chat and a folder. So you could actually open up a different directory and use that as context for the project that you're talking about. Kind of how Claude Cowork works, right? So if you're using Claude, the way that this works is you have a folder open that you talk to and the agent has context in that folder. And that just makes it hyper-specific and related and personalized to whatever you're talking about. So for example, this is a website that we have set up and an SEO project. And so we can do that inside Claude. But we could do the same thing inside Hermes now as well, right? And we can navigate straight to that folder and say, okay, start using the workflows we've set up from Claude. And you can see inside the chat, this is working nicely. So we said, are you working? Hermes got back to us. And we can see the session is active with Hermes right now, which is great. And so we can easily chat between our agents. We can switch between them. Let's just check if OpenClaw Gateway is set up as well. And we'll double check that Claude code is set up as well. So Claude code is very, very responsive. OpenClaw, unsurprisingly, is super buggy. But don't worry about that. It's buggy every time. Aion UI or not, right? Aion UI or not. And then you can see an example of how you can use the skills you built into Claude code. So this is a previous session that we used on Aon UI. And actually built us a website and deployed it to this subdomain, which you can see right here. And so you can switch between Hermes. You can switch between Claude code. You can search for files if you type in at the bottom inside Aon UI. And you can also attach folders and documents. Here's an interesting one as well. So if you go into, for example, Hermes, you won't see the permissions list here. But you will see how much context you've used. So you can see how much context has been used inside the conversation. If you go over to Claude code, though, what you can actually do is switch between permissions as well. So you can switch between auto, default, accept, edits, YOLO, et cetera. And then over here, we have scheduled tasks. Now, the difference is that normally you would just have one agent, right? So you go into Claude code. This is nothing special. You can have scheduled tasks inside Claude code. Or you can have scheduled tasks inside OpenClaw. Not a big deal. But when you've got multiple CLI agents and you can manage all of their scheduled tasks across every single agent, and you can see it in a nice sort of UI like this. Plus, you can keep all of them awake like this too. So that your computer doesn't switch off and then the scheduled tasks don't get missed. It's really powerful, really easy to use too. And then you can enter the task name, the description, the agent you want to use, or the skill, right? So you can actually have some preset assistance here that are built into Aeon UI. And then you can say, okay, CLI agents, I want to use Aeon. I want to use Gemini, right? Claude code. I want to use Quen. I want to use Hermes, et cetera, right? Even Nanobot is down there. I mean, that's the extent to which Aeon UI works with pretty much everything. And then you can add your prompts and you can automate and schedule tasks. You can have this running 24-7. You can switch between your models. The other cool thing about this as well is like if you're using Aeon CLI directly, then what you can actually do over here is you can add new models, right? And so if you want to add like a local model, you could easily do that, right? You could add a custom model here. You could add a new API. You could add multiple different models so you can switch between them. And I think anything that's OpenAI compatible would work with this, right? Which is pretty nice. And then we have our skills inside the capability section. So you can see here, for example, we have global agents and we have Claude code skills too, right? And this discovers everything on your laptop. Also, you'll see this MCP section. So you can configure MCP tools with the agency you're using. Configure speech to text, right? So you can configure the voice input provider and set that up directly, which means basically you can talk to it in your voice and then it'll actually transcribe that into text, which is great. Over here, you can add image models, which is super useful. The way that you need to do that first though is configure the image models. How do you do that? You go to your model section here and then you would add a new model, right? And then you would add more on that as well. So if you want to see it as we're using to make sure that we are looking at that.