← Back to search

Herald OS DESTROYS Hermes Agent Desktop?

AI News Today | Julian Goldie Podcast · 2026-10-08 · 11 min
relevance 100 2292 words Episode page ↗ Audio ↗
Show full episode description
Herald OS DESTROYS Hermes Agent Desktop?
✨ Episode Outline — click any point to jump to it in the episode
Problem solved
Running Hermes Agent from the terminal or Hermes Desktop is messy; Herald OS gives it a polished, agent-native desktop UI.
Benefits
  • Free, open-source Herald OS on GitHub, set up via Claude in ~2 minutes
  • Dashboard with daily schedule, needs-attention items and pick-up-where-you-left-off
  • Separate Work, Personal and Ideas spaces for multiple clients or companies
  • Missions, memory, files, automations and activity history in one place
  • Settings for model, Docker runtime, background agents, auto-approve and voice engine
Use cases
  • Automations synced across Herald OS, Hermes Desktop and an Agentic OS: 6am to-do recap, daily Japanese practice, inbox triage
  • Giving Hermes a mission like 'build out a new website' and it plans step by step
  • Switching Work/Personal/Ideas spaces to keep separate clients cleanly organized
  • Pasting a GitHub plugin URL to extend the agent on the fly
KPIs / results
  • Setup via Claude took about two minutes
Tools / build
0:00 / 0:00
So today we're going to be looking at new free Hermes Agent operating system, an agent native operating system with Hermes Agent as the interface. And I've already set it up. It looks absolutely awesome. Look at this bad boy. Look at the design, the UI, the way it's set up. We'll come on to that in a second. And this is free and open source. Shout out to Luke, the developer. He posts loads of stuff on Hermes on X, but you can see, for example, he's just dropped the Herald operating system and is pretty cool. There's so much you can actually do with this. You can also, for example, make it your own. You can personalize it, but we'll come on to that in a second. And if you want to get it, it's available on GitHub Herald OS, agent native operating system with Hermes Agent as the interface. Now, if you're wondering, how do you set it up? For me, I just literally went into Claude and I said, can you set this up? And within about two minutes, it already set up the agent OS for us. You might say, why would you want an operating system? Basically, it just makes it easier to navigate and control your agents and command them. Because otherwise what happens is that I remember when this was like the first time the Hermes came out and it dropped. I remember we all used to have to use it inside the terminal and that was not long ago. We were looking at maybe March, April or something like that when we were using Hermes agent. And you just operate it from here and it was super messy. Whereas, for example, you've got a beautiful UI and this is the great thing about open source projects. You can use it inside Hermes desktop, which I personally don't think looks as nice as Herald OS. So you've got Hermes desktop here and you can add some really cool plugins to it. Or you've got Herald OS, which I think looks much better, much nicer. So as you go through here, you've got this dashboard and it says, good morning, Julian. Your day already in motion. And then you can plug in what you want. You can also talk to it over here so you can drop in your voice. It also looks at your day. So what's scheduled in for the day. If you look at this, this is actually taken from our scheduled tasks inside Herald OS. And then it's like, okay, here's what needs attention. Working with you. And also you can see here that you can pick up where you left off. So when you open Herald OS or you come back after a break, Hermes looks at the names of your recent documents, projects, chats, calendar events, and suggests where to continue. So you can switch it on or switch off if you want. And you can also give it missions, which is really cool. Now over here, you'll see that it's actually taken some screenshots. So what we've been recently working on, this is actually from my desktop. So you can see here, it's got some screenshots that I've recently taken and then use that inside the dashboard here. If you actually just want to use Hermes normally, you can use it over here. And you've got an option of your preview and sources. So you can add sources and files that Hermes can read. You can start a new conversation over here. You've got the preview. And then you can switch between personal and work, which I really like. So you've got three different options here. You've got a workspace for ideas, which you can just kind of jazz things out. You've got the work section here where you can switch between the stuff that you do for business. And then you have personal. So this is really useful because a lot of people with agentic operating systems, and I've seen this inside the AI Profit Boardroom community, with agent operating systems, I quite often get the question, which is if I have separate companies, or if I have separate clients, how should I set up my agent OS? And this is a really clean way to just set that up step by step. So we can switch between work, personal, and ideas, as you can see. Then we've got the missions over here. And so this is basically where you can give active missions to Hermes agent, and it will plan it out step by step. So we can type in a new mission here, and we could be like, okay, build out a new website for me, and then it will start going off and building for you. Over here, we have the memory. And this is pretty useful because I hope it's not showing any. I can see it's got my telegram details right there. But if we have a look, for example, over here, we've got, yeah, so you just need to be careful what we show there. But basically, inside here, it breaks down all the memories it's got for you and everything else. Now, inside the files section, you can basically manage all the files the Hermes has access to. So we can switch between recent, favorites, projects, downloads, shared, and then stuff you've deleted as well. And then you can see here, for example, recent. These are all the files inside your folders, as you can see. Then we have automations. So these are like the stuff that you set up. You can easily manage them, switch them on, switch them off. The other thing that I love about this is it's syncing automatically between three different systems that I have as I test a lot of these tools out. I'll show you what I mean. So we have Herald OS over here. We have Hermes Desktop over here. And then we also have Agentic Operating System, which we have over here. And inside the Agentic Operating System, we can switch between all our agents, not just limited to Hermes. And so what we can do is we can easily manage Hermes in three different places. And it seems to sync up nicely. So, for example, daily 6am to do recap, Japanese practice, daily Japanese learning, inbox triage daily, morning inbox summaries, etc. It all looks super nice. It's quite easy to manage and measure from there. Over this section, we have the connections. You can also resize the bar as you can see. So you can just move that to the side if you want. One thing I've noticed slightly, and let's just see if we can... Yeah, it cuts off a little bit at the top, but this is an open source project. So who am I to be picky about that sort of stuff? Over here, it seems to have a connection section. Now, that bit doesn't work for me. I think if I wanted to fix it, I could easily just go into Claude and say, hey, can you fix the connections? Can you have more apps and stuff like that? So this is basically where you can bring in tools, bring in apps, bring in external connections directly to your AI agents. This is pretty useful, but as you can see, it's not synced up for me. And this is one of the things about open source projects, again, is it might take a little bit of back and forth for me to really get this working and synced up nicely. But it looks cool. And then we have the work section, ideas, and settings, right? Now, inside your settings, you can switch between what's your preferred model. So I've got GP6 Astra with the CLI plugged in over there. We can decide where our agents complete tasks. And here you can see that we can actually switch between different setups. So a lot of people say, I want to run Hermes on Docker. And you can switch it over there. You can switch it to these different ways to run it. And then we've got background agents. So this is like how many agents can run in the background. We can allow it to remember our preferences. You can also allow auto-approve on and off. So ask before sending messages. And then activity history so you can see what happened here, which is super useful. Because if you've got multiple devices, multiple things going on, you definitely want to see what's going on inside your history. Because if Hermes is working on something that you've forgotten or something that maybe is running in the background that you didn't even know about, it's pretty useful. And believe me, like when you're messing around with stuff like Hermes, lots of stuff can get set up in the background that you totally forget about. Now, inside the general section, you've got do you want to launch this full screen? For me, not really. Don't really like to go full screen. Default space. So here you can switch to in personal work ideas. If you want to add a client or something like that, you can just add in a new space. Like you can decide what's your default folder to work from. Projects folder. So here you can switch to in do not disturb mode. So that basically just switches off the notifications. Because it's actually got a bell notification sound that will pop up if you're running through this. You can decide what stays awake or not. Screen saver. Then you can switch between dictation. So do you want to allow dictation or not? You can quit the session if you need to. And there's all these cool sections. For example, over here, you can change the colors that you work with. I actually quite like that one. I think the default is pretty nice. We can choose which software we have access to. So these are coding agents and local models in the background. And we can remove them. We can actually manage them from this section here. So for example, we've got Call Code, Codex, LM Studio, etc. And then we have the plugin section. So here, what this means essentially is if you see something cool on, for example, GitHub that you like the look of, you can paste it in here and then click add. And just modify your AI agent as you go along. Then we have the Hermes agent section here. We have the usage section so we can see how many tokens we've used, what plan we're on, the token balance, etc. Voice, probably one of the most useful parts here. So you can switch the engine that your voice runs on with Hermes agent. So number one, that means you can speak to Hermes just like you would with, I don't know, Jarvis or something like that. But you can also switch between how fast it is. So for example, over here, you can use a local voice agent or you can run with the live one. You can try and test it out. You can have a look at the commands here as well. So there's all these different voice commands that run with Herald OS. There were the privacy, notifications, switch to networks, storage, accessibility, etc. But super nice, super clean, easy to use, synced up pretty quickly. And a great option, I think, for those people who are looking for a nice UI, but don't really like Hermes desktop, for example. So that is basically it, my friends. That is Herald OS. It's free. It's available on GitHub. You can get access to it right now. There's a lot more to it. If you want to see me do more videos on this sort of stuff, feel free to let me know in the comments. I'd be interested to see if you want to see more things about that. And by the way, if you want to get more training on Hermes Agent, how to use it, and a community of people using Hermes Agent, you can get that inside the AI Profit Boardroom. Link in the comment section or just go to the AI Profit Boardroom.com. Inside the community, you can post questions, get help and support in real time, and I'll personally answer them. I often create personalized video tutorials for people just to help them as much as we possibly can. So, for example, James asked a question about lead generation yesterday, and we already created a personalized video tutorial for him just to help him as much as we can. Inside the classroom, you can ask questions, and you can get access to all of our best trainings here. If you like Hermes Agent, we have a lot of new daily updates on Hermes Agent on this section. So we're always adding new tutorials, new step-by-step guides, et cetera. And then also, we've got courses, for example, on how to build an agentic operating system, how to use Hermes Agent, how to use GFAI, Claude Opus, et cetera. So we're always adding new courses, and this is always evolving to help you. Inside the calendar, you can drop a weekly coaching course, ask questions, get help and support in real time. And in the map, you can meet people in your local area who are building out with AI agents like you. So feel free to get this link in the comments description or go to the AIprofitborn.com. Plus, you can personally message me inside there too. Cheers to watching.