Agent OS: GPT 5.6 + VPS + FREE Apis!

So today I want to be answering some of the latest questions inside the AI Profit Boardroom. This is something that I do every single day inside the community. If you have any questions yourself, feel free to post. Link in the comments description.

And I answer these questions and just try and help you as much as I can. I know if many of these people have the same questions, then you probably have these questions too. So let's get straight into it. So Alex is saying, and he's talking about the Agent Operating System, which is a system where we basically have a mission control for all of our AI agents.

So we can put in our goals, we can journal, we can see all our agents working online and everything else. We have them all plugged into one system, right? And it works with a memory system too. And so this is something that a lot of our community members love about the community.

So the first question that we have from Alex is, I've been building and expanding my operating system and was wondering if I could set up the folders in iCloud or on a drive that I can access somewhere and I can use it across my five Macs. So I think one of the easiest ways to do this would be to get your Google Workspace API. And with Google Workspace, you can give your agents an API that they can write files to. Now, if you share access across your OS on multiple Macs, then any files that you create on Google Drive can be shared with the same access token across every single Mac and every single instance of the operating system as well.

Another option, second option, is that you have a VPS setup and then all of your agent OS instances run directly inside that VPS and you could save files locally on the VPS. We actually have a few tutorials on how to set up an agent operating system with a VPS below. You can use Cloudflare to set that up with Hostinger. So two options there.

Number one, you set up Google Drive and you share that across all your agents. So for example, if you've got Hermes agent or Codex working on multiple different devices, they can all save to Google Drive. Or the other option is that you have a VPS and you save files locally. We got another question from Jason.

So Jason was asking like how to use the latest new ChatGPT app. I've actually done a one-hour course on this and put it inside the Aircraft for Boredom already. But basically, Codex is rebranded to ChatGPT Work. And with ChatGPT Work, you have ChatGPT built in.

So you can click on the chat over here or you can use the Codex features inside here as well. So if you want to build stuff, use the projects on the left hand side. If you just want to use ChatGPT inside there, you can use the chat here as well. And if anyone's running on the older version of Codex, just update.

You'll get the latest version with both of these features. But yeah, it's basically like one big super app that can build apps, websites. It can create videos for you. It can connect all your favorite apps and it can create schedule tasks as well.

Really, really good. I also prefer the UI inside ChatGPT versus Cloud. It seems a lot easier to use and GPT 516 sole itself is absolutely amazing. If you want access to the system with the one hour course, here it is.

But there you go. Next one. So this is pretty cool. Connor is saying like, I would like to create an AI avatar in order to create a training course for team members.

Also to be able to create one-off videos and everything else. What are the right tools to set this up? So we actually have a full training on this. And this is the same system that our whole team uses.

So you can see it over here. And this is like the AI avatar setup. Now, if you want to, for example, add nice like B-roll and stuff like that, we actually show you how to automate that over here as well. So if you want to set up the voice too for the AI avatar, we have a full process here.

But yeah, AI avatar videos are super powerful. I've seen a lot of people get great results with them. As an example of that, if you actually go over to our Twitter account, you can see like, we really only post like AI avatar content on there. And yesterday we got like 550,000 impressions in one day.

So you can reach a lot of people. And a lot of people actually prefer, sadly, we live in this world, a lot of people actually prefer our AI avatar content. So yeah, that's it. And then you can see like loads of people inside the community help each other as well inside this group.

So there's lots of good tips inside here. For example, let's have a look. Sheena actually just built a learning platform for a client that includes AI avatars. David was using Gemini for this as well.

So you can use Google Gemini, but loads of great tips inside there. This is really cool. So Lee is working on a AI employee platform where basically he can give his agents missions and they work creatively to solve the problem. And he's actually shared the whole process.

Look at that. How cool is that? This is some of the stuff people working directly inside the AI Profit Boarding, like some of the cool projects. I think it's a great example because this is great.

Like, number one, thanks so much for sharing. But number two, how crazy is that? That you can build out your own platform and have autonomous agents just go off and do work. Absolutely mind-blowing.

So we've got a question from Trevor here, which is, how do you use the free API, which is OmniRoot? Actually, and then, you know, can you give Hermes Hermes workflows with OmniRoot as well? So when we're using OmniRoot, there's a few options. Some of the ways that I use OmniRoot are, number one, with the OmniRoot coder inside the agent OS.

Number two is with free cloud code. And number three is with Hermes. And it seems to work okay. If you're getting the fetch failed, I would just make sure that for whatever agent you use to set up the agent OS, you give the details of OmniRoot and the GitHub directly to the agent and say, keep testing it until it eventually works.

Because it might be that the agent that set up for you has routed it incorrectly and just needs to iterate. But anytime I have a problem like this, I always just give it to the agent and ask it to fix it and test it for me until it finally works. And then I just give it a final check. And if anyone's not familiar with OmniRoot, it's a free GitHub where you can basically code for free across like 90 different models.

And it's like a free API for coding. Obviously, it's not like Fable 5 level, but it's pretty good. And then, for example, if you wanted to set this up with Hermes agent, let's say, for example, I'm using ChatGPT for work, I can go into the Hermes agent project, which we have over here. And I can say, create a separate profile inside Hermes agent that uses OmniRoot, then test that it actually works and iterate until it finally works.

Make sure it's all configured properly inside the agent OS. So that's how I would approach it. Now, there's a good question from Matt here who's asking about OMI. Now, if you're watching this, you're not familiar with OMI, basically, it's a way that it can listen to your conversation.

So it can listen to my microphone and it can watch my screen. That means that it has a lot of memories on me when I'm working on. So for example, you can see 45 minutes ago, here's what I'm working on. And it like takes notes on me and what I'm working on day to day.

Now, one of the problems is that it's fantastic for using inside Obsidian as a knowledge graph. But one of the biggest issues is that it is not running on Windows. So how would you get that set up? There's actually an alternative to this, which is another free GitHub that you can use instead.

And you can use it on Windows. Then you can ask that GitHub project to sync with Obsidian instead. And you can use the plugin system with the AI prompt journal. So if you're on Windows, you want to use OMI, you can use it like this instead.

And this basically does the same thing. So that is basically it for all the latest questions inside the AI Profitable Boarding. We've handled every question, given instructions clearly on how to like solve each problem as well. If you want to ask me questions like this, feel free to join the AI Profitable Boarding.

It's just an awesome community where we win, we learn, we grow together. You saw all the awesome wins people share. And we actually have over 197 pages of testimonials from people winning, learning and growing with this sort of stuff. So you can see all the awesome reviews and testimonials.

It's just a great place to learn and share everything together. And I think as well, when you're learning AI, like the fastest way to learn and get the most out of it is to be part of a community where everyone's just leveling up and improving every single day. So if you want to get this, feel free to get it inside the AI Profitable Boarding, link in the comments description. You can ask questions inside the community.

There's always people online 24 seven as well. So you can get help and support whenever you want. Inside the calendar, you can jump a weekly coach calls, share your screen, ask questions in real time. Inside the classroom, you can actually get all of our best training.

So if you're a complete beginner, you can go from beginner to expert over here. And if you want to get our new daily updates, we have them inside here. So for example, we've just added a new one hour course on how to use chat, how to reduce the amount of tokens you use, how to use free APIs like tensor and HY3. And also you can get a full agent operating system inside there too.

Inside the calendar, you can jump a weekly coach calls, share your screen, ask questions in real time. And then inside the map, you can meet people locally who are building with AI agents like you. So feel free to get that link in the comments description or just go to the AIprofitboarding.com. Thanks for watching.