New Hermes Agent OS is INSANE! artwork

New Hermes Agent OS is INSANE!

AI News Today | Julian Goldie Podcast

July 16, 2026

New Hermes Agent OS is INSANE!
Speakers: Julian Goldie
**Julian Goldie** (0:00)
This new Hermes Agent Operating System is absolutely wild. We've actually got GPT 5.6 plugged into the system right here. And today, I'm going to talk you through the Hermes Agent Army, which is a powerful system where we've got, for example, Hermes Apollo, Hermes Oracle, Hermes Astrost, a chat over here. We have Goal Mode plugged in. We even have an email outreach tool with Hermes Agent, and this all connects to our whole system. So, for example, if we want to generate AI films, we can do that over here with Open Montage. If we want to edit videos, we can do that with the video editor. If we want to use open code, we've got that plugged in over here. If we want to generate images, we've got that inside this system. So we can pretty much automate and build anything using the Agent OS. And today I'm going to talk you through the Hermes Agent Army. So you got one command and a whole army of agents. And the thing is, most people run in AI like one chatbot. Inside the Agent OS, Hermes is an army. We've got over 30 profiles, 14 different stations, voice command, 24-7 radar, outreach and SEO squads. And every mission actually saves inside my memory vault. So we've actually got this memory system where everything that we do with our AI agents is plugged into the system. All of our other agents, Claude, everything else gets updated automatically using our Obsidian Vault. And this is all plugged into this amazing system right here. So let's talk about why we do this and why this matters. So back in the day, you know, I don't know about you, but if you're using, for example, Hermes desktop or Hermes terminal, the problem is that you are the only employee of your AI, right? So you'd have like one chat window and you're basically the secretary. Which means every job, you're typing, you're waiting, you're copying and pasting the answer somewhere. They're starting the next one. Research, outreach, SEO, video. These are all like one at a time workflows typically. And so previously for me, my AI was smart, but it was just one worker and I was its manager, its courier, its filing clerk. I was basically the employee running the whole team. And then I turned one Hermes agent into an army inside my agent. So now one command line reaches over 30 different AI agents. I've got a radar that would show my competitors, whilst an average squad fills my pipeline with new leads. I talk to it out loud whilst it builds, missions run in parallel whilst I sleep, and every exchange order saves to my vault. So the army never forgets the thing. And this is the same agent you can get access to. It's just Hermes, but it's how it's organized that really makes the difference.
So, let me show you an example here. If we go into Hermes agent, I've got, for example, Hermes Apollo plugged in, and with Hermes Apollo, we can actually build and speak to it in real time. So if I say, hey, how's it going today? You can see that it actually replies. I'm not going to play the sound because I don't think it will sound very good on this video, but you can see that it's replying over here and it speaks back to us. We can also see our full transcript of what we've been talking about over here, and then we have everything that we've built with it. So you can even build websites with my voice, and I just give it the command. That is unbelievable, and 99% of people don't know how to do that. We can also have it in wall mode, and we can speak to it. It can wake on a wake word. We can have it run in the background. The same, for example, with Hermes Astros. This is monitoring my compared to 24-7. So this is the sort of stuff we're talking about here. And you might say, well, that sounds technical. How do you get this set up? So we've actually got the AR Profile Boardroom, where you can get the Aging OS system from us, but we've got over 200 different wins and testimonials from this. So these are all non-technical people. I'm non-technical myself. I'm not a coder or anything like that. And the great thing about this is that if they can do it and I can do it, you can do it too, right? Whether you build it yourself or whether you use our system. And so let's talk about why this is so important. So the problem here is like, you've hired a genius and you've made it a typist. And here's the problem with how almost everyone runs AI, including me, for far too long, which is you've got the smartest employee in history, right? Hermes agent, but it's just sitting in a chat window, sitting in a desktop app. And its entire job is waiting for you to type. It can't start anything on its own. It can't watch your competitors over night. It can't send the outreach, plan the cluster, build the page whilst you're on a call, because the moment you stop typing, it stops working.

11 more minutes of transcript below

Feed this to your agent

Try it now — copy, paste, done:

curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Works with Claude, ChatGPT, Cursor, and any agent that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire.

Using your own key:

curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000777119021