← Back to search

Claude Agent OS is INSANE! 🤯

AI News Today | Julian Goldie Podcast · 2026-08-03 · 10 min
relevance 85 2350 words Episode page ↗ Audio ↗
Show full episode description
Claude Agent OS is INSANE! 🤯
✨ Episode Outline — click any point to jump to it in the episode
Problem solved
Fragmented AI tools that don't share memory or context, forcing constant model-chasing and manual repetitive work.
Benefits
  • Runs 24/7 in the background while you do other things
  • Shared Obsidian memory system so new models need no retraining or new prompts
  • Swap models in/out (DeepSeek, Claude Opus 5) without breaking automations
  • Built for free using existing subscriptions, CLIs, and free APIs
  • Auto-notes daily activity into a linked knowledge graph
Use cases
  • Hermes Oracle pulls last 24 hours of news daily, then one-click social/SEO content
  • Hermes Astros monitors competitors and keywords, delivering 6 personalized content ideas/day
  • Video agent fully auto-produced an AI avatar video (B-roll, voiceover, edited) about Minimax H3
  • Integrated Higgsfield via MCP as an on-the-spot film studio for promo videos/images
  • Wake-word 'Hey Hermes' voice agent (Hermes Apollo) running on a wall monitor
KPIs / results
  • Runs 24/7 with no extra cost beyond existing subscriptions
  • 6 personalized content ideas generated per day from competitor monitoring
  • 207+ pages of member wins/testimonials in AI Profit Boardroom
  • ~3-4 hours daily spent improving the system
Tools / build
  • Claude Agent OS (mission control dashboard)
  • Hermes Apollo voice agent with wake word
  • Hermes Oracle daily news-to-content pipeline
  • Hermes Astros competitor/keyword idea monitor
  • Hermes Mixture of Agents multi-model orchestrator
0:00 / 0:00
This new Claude Agent Operating System is the most powerful thing I've ever built. So this is a system where all of our AI agents working together. So you can see for example we have Hermes, we have Claude, we have Anti-Gravity and this is all running 24-7 in the background whilst I do other things. So what this actually does is it links to our memory system here which means anytime a new model comes out we don't fall behind, we don't have to retrain it, we don't have to plug in new prompts. It's already got all of our context pre-loaded using this free system which is basically Obsidian. Now also when we're running this we have an automatic system that takes notes with our AI agents. So for example I wasn't even online an hour ago but this takes notes based on what we do each day, plugs in all the notes into our memory system and then if we go into our knowledge graph you can see that everything is linked here. And each one of these little dots is a memory that our AI agents can understand. Now over here we have Claude and anytime a new Claude model comes out, so for example like Claude Opus 5, we've plugged that into our system so that it's ready to go and ready to use whenever we need to. And it's automatically updated. Over here we have Hermes Agent and as soon as something new or useful comes out, for example like OpenCode just released DeepSeek0731 as a free API inside their system, we can plug that in as a free API inside our Hermes Agent. The same for example with Hermes Apollo, this is a custom built workflow which is a voice agent we can talk to and it can talk back to us and it's basically like having a real-time Java so it can speak back. We've even got for example this whole system here where we can build things in real time and we can view everything we've created and then we can come back to it. And I think the main benefit of that is that normally when you're using a system like this, you don't need to worry about, okay where are all my files? Well, where's that project I created like a week ago? It's all there ready and saved inside one system. We also have the wall mode here. So what this means essentially is like we can have everything saved in one place and we can have this running on a monitor in the background so that our voice agent is ready to go whenever we want. And we can actually use a wake word. So we have a wake word system here so that it's voice activated and we can switch it on just by saying hey Hermes for example. Over here we have DeepSeek with Hermes Oracle. Basically what this does is every 24 hours it consults the Oracle on what's going on. And so we have for example this system here where we pull in the latest news from the last 24 hours. It runs automatically and then if we want to create social media content about that trendy topic we can click here. And if we want to create SEO content we can click here. Now there's a few benefits to this. Number one, it runs 24-7. Number two, we can create social media content automatically with it. Number three, we can create SEO content automatically with it. Based on trending topics. And number four is that we never fall behind on industry news again. We also have Hermes Astros over here. And what Hermes Astros does is it pulls in the latest ideas from our competitors and keywords that we are monitoring. So if we go to our watch list you can see that we can monitor our competitors. And we can also monitor our keywords. And we can change or add new keywords at any time. And then we get six new ideas a day that are pulled in from our competitors. And are basically personalized to us. They give us personalized ideas in terms of what we can actually create content around. So if we click on this for example we've got the idea over here. And then we also have for example the unique titles for us which are generated by Hermes Agent. We've got the angles and we can click that into our video agent. Into Notebook Lem. Into our SEO content system. And the great thing about this is that it's basically personalized to us. It means that we can automate social media content. But also we can do it in a way where it's fully customized to us. Our industry, our keywords, our competitors as well. And you know it saves a lot of time on research. We also have the studio over here. So that we can plug in Minimax or Grok with Hermes to generate videos or voice notes or images. And we can manage all the sessions that are running in the background there. We also have something called Hermes Mixture of Agents. Which basically combines multiple models with Hermes. And by the way this whole system that I'm showing you today was fully built with Claude. So I can just use my existing subscriptions with Claude Code to build this out. Now the other great thing about this is when we're doing this, you know, it doesn't cost me anything extra. Because I can plug in my existing subscriptions. So for example with Hermes I can use free APIs or I can use CLI so I'm already using it. With Claude I can plug in my CLI. With Antigravity you can plug in the Antigravity CLI, right? And so if you want to use this for free you can do that right now. And you don't need to worry about it at all. Now also it's the same with the memory system. The memory system, Obsidian, is free and open source. And so it's free to use a lot of this stuff. For example like even Open Design or Open Montage. These are free open source projects that you can use to build and automate anything whatever you want. Which is pretty cool. And then the other thing that I like about all of this system is that you customize it to whatever you work on day to day. So for example if I spend a lot of time doing video, guess what? I'm going to automate that and create a custom workflow where I can just type in a prompt and generate a video on the spot. So if we have a look for example at this system. We actually generated this video with AI avatar, the B-roll, the voiceover. All fully edited and put together like this. About one of the latest updates which is Minimax H3. And it was fully automated using our video agent. And I don't want to say this to boast or to show off. But just to really inspire you in terms of what's possible. Because most people are like okay well where do I start with AI? How do I get the most out of it? What are the use cases? These are the use cases. What you do day to day is the use case. Because you can automate that. Anything that's repeatable you can automate with it. And you can also save a lot of time here. Because you're not clicking into Claude and then clicking into Antigravity. And then opening up a terminal. And then opening up ChatGPT. And using all these tools that don't even sync together in terms of memory and context. You can just use a system like this. And it's way more powerful and easier to organize. The other thing that I like to do as well is like anytime something new and useful comes out. For example like DeepSeek. We will plug DeepSeek into a system like this where we can use the latest models. So we don't have to chase models anymore. We don't feel like we're falling behind anymore. We've got a system that improves every time a new model comes out. Not goes backwards. And we don't break any automations every time a new model comes out. Because we can simply swap in and swap out the models that are new and that are old. Also over here we have for example new systems that we can build on the spot. So for example I actually integrated Higgsfield into our system here. So that we've got a full like film studio ready to go. I can for example create promo videos or beautiful images like this on the spot. And I can also pull in the latest video models from Higgsfield as well. Which is awesome using the MCP with Hermes. And again that was built on the spot. And if you're wondering like how do you build this sort of stuff. You literally just go into Claude or whatever agent you prefer. And you say listen build this. And then you test it and then you iterate it. And if you're wondering what to build. You focus on whatever you're doing day to day. If you're not sure what you're doing day to day. You just time what you work on. And then you make a list of the tasks that are repeatable. And then you start automating those. The other thing that I would say here. Is like usually I'll spend about three to four hours improving. And making this better every day. And then I share it with my AI Profit Boardroom community. So if you want to get this link in the comments description. Or just go to the AI Profit Boardroom dot com. And also what we've got here is a mission control dashboard. So this just helps me see the status of each of the agents that we use. I can also plug in my goals list here. And I can see my progress of goals over time. Because I can see my history. And what I created previously and did previously. Which is super useful. And then we can see previous logs of what's going on over here. And we can also plug in new features. So for example of Claude. When it came out with UltraCode. We actually plugged that into this system. So that we can run multiple sub-agents. And orchestrate our agents together. Now some people ask me inside the AI Profit Boredom. They're like Julian. How do you tag AI agents. So that you can just at sign them. And then get them to talk to each other. We actually created this system here. Which you can see where basically we can say. Hey Hermes. Come up with a new idea for an app. And then our agents jump in. And once something actually useful comes out. It'll actually take a note of it. And plug it into our pipeline. So that we can create that. And get that working automatically. So it's a powerful system that just gets better. Improves all the time. Super personalized. And also you can just set this. I mean you could code this yourself. But if you just want to get my version. You can get it inside the AI Profit Boredom community. Link in the comments description. Or just go to the AI Profit Boredom dot com. This is my AI automation community. For learning, growing and saving time with AI automation. The beautiful thing about this is. You know a lot of people say. Building an agent operating system. Or using an agent operating system. Not. It's actually not. So I am not technical at all. I can't write HTML or code or anything like that. But I can build a system like this. And we have over 207 pages of wins. Of testimonials. Of people learning, growing and sharing what they're doing with AI. Inside the AI Profit Boredom. And people are crushing it with this. So for example Kareem built out their mission control. And it looks absolutely amazing. Andrea says. Having my mission control OS has been a game changer. You can see for example. Rich said. Does anyone else blown away by the AI Profit Boredom. And so so many people are getting amazing results of this. That I know that if I'm non-technical and I can use it. And they're non-technical and they can use it. Then we can all build and grow with these systems every single day. So if you want to get this link in the comment description. Or go to the AI Profit Boardroom dot com. Now inside the community. You can ask questions. Get help and support inside the community. Day to day. And inside the classroom. You can actually get all of our best training. So inside the community. I personally answer all of the questions here. Every single day. And also create a video tutorial. Running through all these questions. To help you as much as I can. Inside the classroom. You'll actually get the agent OS. Inside the new daily update section over here. So you can grab the agent OS system. We've got a full video tutorial. Showing you how to install it. You see when it was last updated. You get the zip file to install it. And then every time something new. And useful comes out. We show you exactly how to use it right here. With a step-by-step guide. And a tutorial. Inside the calendar. You can jump on weekly coaching calls. Get help and support in real time. Ask questions. Meet other members. And that's live on a Zoom call. And then inside the map. You can meet people locally. Who are building with AI agents like you. So feel free to get this. Link in the comments description. Or go to the AI Profit Boardroom. And then inside the map.