← Back to search
Hermes Agent OS is WILD!
AI News Today | Julian Goldie Podcast · 2026-10-11 · 9 min
Show full episode description
Hermes Agent OS is WILD!
✨ Episode Outline — click any point to jump to it in the episode
Problem solved
Using
Hermes Agent through a bare desktop is boring, breaks and overwhelms; an agentic operating system UI automates and personalizes every part of Hermes.
Benefits
- Run Hermes by chat and Hermes Apollo voice at the same time
- Plug in free models like Step 5 via OpenRouter for coding
- Hermes Oracle auto-scans trending news every 10-12 hours with content angles
- Shared Obsidian memory keeps Hermes, Claude and Antigravity in sync
- Custom SEO tool finds ranking wins, content gaps and generates articles
Use cases
- Hermes Oracle pulls Grok trend data, scores topics, one-click turns them into posts or WordPress publishes
- Hermes Muse rescans own content daily to show what is popping off and what to create next
- SEO tool: 30 days of search data, content gaps, one keyword plus case study yields five deployable articles
- Hermes Astros analyzes competitors, feeds a video agent that scripts and builds AI avatar videos
- Free AI coder with Step 5 builds and saves mini apps as a sub agent
KPIs / results
- Hermes Oracle rescans roughly every 10-12 hours
- Five new articles from one keyword and one case study
- Creator spends about three to four hours a day coding the system
- Step 5 free on OpenRouter for the next week
Tools / build
- Hermes Agent OS UI with profiles and mission control
- Hermes Apollo pop-up voice agent (Codex voice or API)
- Hermes Oracle trend scanner (Grok)
- Hermes Muse and Hermes Astros
- Obsidian shared memory with create tab
Today I'm going to show you a powerful system that I've set up with Hermes and this basically automates every single part of Hermes but also presents it in a beautiful UI which is what you want when using Hermes. I will tell you now, using an agentic operating system like I'm about to show you now is probably the most powerful way to use Hermes Agent and also you can plug in free models. So for example, we've just plugged in Step 5 which is a new free model we can use with Hermes Agent. Super fast, really awesome to use. You can get it free for the next week. And additionally, we've also plugged in Hermes Apollo as a pop-up voice agent. So what this means essentially is we can use Hermes inside the chat here, we can switch between different profiles, we have loads of different profiles built into the system as you can see. But then also, we can also use and speak to our AI agent over here. So we can use two different versions of Hermes at the same time, one via voice and one via chat. Now, if we go into Hermes Apollo here, this is like a pop-up version and it stays persistent as we switch around, right? So that Hermes Agent voice mode is just ready to go at every single step of the process. I can then close it or end it, I can minimize it if I need to, I can change your voice on that as well. So if we click over here, we can change your voice and I can say, hey, how are you? Very well, thank you sir. How may I assess? Do you know what's been going on in Obsidian recently in my memories? Just a moment, sir. I'm checking. The latest notes mention questions about Agent OS, but I don't appear to have your latest... So you get the point. And then also you can see over here, what we have is this section called Hermes Oracle. Now Hermes Oracle, basically, it's analyzing the latest trends in our industry. It's finding the latest news. So you can see it's pulling in here. And also we can see when it was last updated. Now it's re-scanning every, I would say every 10 to 12 hours, but you can scan it as much as you want. So it always consults the Oracle, tells us what's going on. We can also listen back to the brief if we don't want to wait. And then we can see all these different pieces of news here. So we can actually click on these and see the individual trending news that's dropping. Now it's also scored by what's trending right now. So which topics are the most important, the most interesting? So you can see right now, this is the most important and interesting topic that's just dropped. And then additionally over here, we can also quickly turn this into a post or we can publish it into WordPress. So what this means essentially is I can create SEO articles, I can create Twitter content around this particular topic. And it even gives me an angle for the content I'm going to create. Now it pulls in the latest data from Grok, which Hermes Agent is really good to do. If you're already subscribed to Twitter, you can just plug in your existing Grok subscription and then you're good to go on there. Now over here, we have Hermes Muse and Hermes Muse is basically looking at my existing content and telling me exactly what content is popping off right now. What is hottest content right now? So we know exactly what to create next time, right? So basically, this is analyzing all of my current content. It's re-scanning on a scheduled task every single day. And then over here, it's pulling in the latest trending topics and what we should create content around next based on what's already worked. Now, we also have cool workflows like Hermes Astros, which basically analyzes the latest trending news. And then if we go to, for example, Apollo, this is our voice agent in full screen mode. So we can see all of our conversations. We can change the voice. We can switch between codecs or the API. Personally, I prefer to use the API. And then if we click into it, we can start talking. Hey, how are you? At your service, sir. What can I do for you? Hey, could you just give me some ideas on like what happened in the AI industry today? Certainly, sir. Allow me to check that for you. And then we can just pause that for now. But you get the point. So we can also type to it and talk to it. And then it can build stuff for us as well. So for example, like this actually crafted a full web page for us just whilst talking to it, which is really cool. So we've got the chat here. We've got the Hermes bot screen here, which is like different AI agents you can talk to. We have Hermes Apollo, the voice mode. And that can build stuff and we can look through conversations and that sort of thing. Then we have Hermes Oracle with the pop-up voice mode ready to go whenever we need it. We can minimize that. We can add it open here. And that's connected to codecs as well. So it doesn't cost us anything extra if you've already got chat GPT because you can use codecs voice from that to plug into your system here. So over there as well, we have Hermes Muse and then we have Hermes Astros. What Hermes Astros is really good at is analyzing our competitors and seeing, okay, what's trending right now? And then we can plug that into our video agent and we can just plug this in. It can write the script, create the full video for us. And if you go to our workspace, you can see examples of videos that we've actually created. It's very good at the motion design. It's very good at avatar videos as well. So you can see some examples here. And then also additionally, it can create full AI avatar videos of us and use our voice and FNL. So you can see an example right here. Fusion fans, your prompt across. It looks really nice. It's added the B-roll. It's created the full video. It wrote the script, put it all together. It looks beautiful. One of polished system. Now also we have, for example, like Claude over here. So we can chat to Claude whenever we want to. We've got anti-gravity ready to go. And one thing that I really like inside the system is the SEO system here. And this is just an example of what you can automate. So for example, let's say you spend a lot of time on SEO, which I personally do. I can basically create my own custom SEO tool here inside the agentic operating system. I am only showing this to inspire you, by the way. Like just to show you what's possible because most people, they're just sitting around in Hermes desktop, just like looking at this, right? And it's no fun. It's not that interesting, right? A lot of things break as well inside here. And it's super overwhelming. The same, for example, for Claude, right? If you go into Claude here, like it's a little bit boring to play with. It's not that interesting. It doesn't make using AI agents that fun. With an agent OS system like this, you can personalize it. You can make it your own. You can change everything that you want here. So for example, I can find my next ranking wins. I can find what keywords we're not already ranking for that we can. We can get all of our search data across the last 30 days. And we can easily research and get the keyword data as you can see right here. Really powerful. And then we can find like content gaps. So if we go to the content gap section here, we can have a look at our keywords and what we could actually create new content around because we haven't created content around it yet. Then if we go to the generate section, we can plug in one keyword, one case study and get five new articles out. And that's basically ready to deploy to our websites automatically. I don't have to log in or anything like that. But we do have a manual quality control system as well inside here if we want to use it. Then we have the video agent. We have the Hermes agent loop system. One thing that I like as well is we've got the free AI coder. And for example, like step five just dropped and it's free on Open Router. So you can actually plug that into this system here where you can code on the left hand side with the free API. You get your output on the right hand side, which is the app you just built. You can save builds over here and come back to them whenever you want. And then also it's got the conversation history. By the way, one thing to note with step five in these free APIs is they look great, but they're not frontier level, right? So it's not going to replace Opus 5.5 or something like that for you. But it will help you just code for free. You can use it as a sub agent. You can use it for coding our mini apps. Most of the time you don't even need a frontier model. Now over here we have the mission control system. So this just gives us an idea of like, what do we need to look at? We can give our agents missions. We can see what agents we're currently using and we can go to our control rooms directly. So if we click on Hermes, boom, shakalaka. We go straight to Hermes. We can change the model over here whenever we want to. We even have lead generation systems as you can see. And then if we go down to the memory system here, this is super useful. So when we're using Hermes, it self-improves. But those self-improvements don't go over to Claude or Antigravity automatically unless you have something like this set up. So this is using Obsidian. Obsidian is an open source memory system where you can basically see your recent memories, which you can see right here. That gets automatically added to the Obsidian database. And that's shared with all of our agents so that all of our agents stay in sync and they have a shared memory. Now what we can also do is we can go to the create tab here. And based on our recent memories, we can create an app, an SEO article, a social media post, a video script, a guide, whatever you want to. This is actually a full video that we can generate with a video agent. And that's all ready to go as well. So it's just, it's super fun. The possibilities are endless. I show it to inspire you. For me personally, I spend about four to, yeah, I'd say about three to four hours a day coding this out. But if you want to get it ready made, you can get that inside the AI Profit Boardroom. Link in the comment description or go to the AI Profit Boardroom.com. If you go to the classroom over here, this is my AI community for learning agentic operating systems. And if you go down to the agent OS, we just added a new update of all these cool new visuals and all the new systems I've shown you. We answer all the FAQs here. So if you're like, I don't know where to start, I don't know what to do. Answered everything with video tutorials and examples. And then we have the zip file ready to go over there. And every time something new and useful drops. So for example, a member actually requested building a system where you can easily understand how to build Obsidian and AI agent memory. We've got that section here with a full guide, as you can see, plus prompts. And then we show you like how to set up a Hermes VPS or how to use OpenDots or OpenMute, OpenBot, which are free open source alternatives to Grot Bar and everything else. We'll also show you some awesome stuff based on what's actually useful and working right now. And I test all this stuff as you've seen today. Now inside the community, you can post questions, get help and support in real time. I personally answer your questions just to help you as much as I possibly can. And then we actually have right now a build challenge. So you can see that we actually have this challenge right now where you can submit your builds and then you can share all some stuff, which is, I think, fantastic. Brings the community together. Now inside the calendar, you can jump all your coaching calls, ask questions, get help and support in real time. And then in the map, you can meet people in your local area who are building out with AI agents near you. So you can just zoom in on the map and DM people near you. So feel free to get this link in the comment description or just go to the AI ProfitBomb.com. Thanks for watching.