Claude Fable 5 Just Changed AI Forever...

Today, I'm going to show you how to use Claude Fable 5 directly with Hermes Agent to orchestrate both of them and get the most out of both of these powerful models. Now, the cool thing about Fable 5 is this is the most powerful model ever released from Claude. It's basically Milfos Class. Basically, you also have Hermes, which is a powerful AI agent.

Now, how can you combine them together? How can you get them working together? I'm going to show you exactly how in this video today, including some of my favorite use cases. So let's get straight into this, and let's talk about how we're orchestrating Claude 5 with Hermes directly.

And I call this the Olympus Protocol because you have two minds on one operating system, and you've got Claude 5, which is probably the deepest thinker you can get, and Hermes is one of the most powerful AI agents for actually implementing its ideas. So if you wire them together into a powerful operating system, they can build, render, design, ship as a team. So let me walk you through some of the most powerful case studies that we've built here. And bear in mind, Claude 5 is great for coding, for thinking, for coming up with ideas, and Hermes agent is great for scheduling tasks, for getting stuff done, for getting it implemented.

So let me show you exactly how this works step by step. So the first one we have here is an agent group chat between Claude and Hermes, and we said, Hermes, give us three scroll-stopping visual concepts for an interactive AI agent explained in 30 seconds as an experience. And then Claude and Hermes worked together to come up with that idea. Then we said to Claude 5, because we've got Claude Fable 5 plugged into this system, pick the winner, give me the build spec, and a cinematic image prompt for the Hermes studio.

And you can see here, it's come up with the ideas and worked directly with Hermes to pick this idea. So we opened up a group chat with two of them, gave them a brief, Hermes came back with some ideas, and then Claude actually rejected two of them, and from there decided to implement one. And then we actually sent the idea over to our pipeline. So you can see the idea that came from Hermes, we've got it right here, ready to build.

So you've basically got two AIs working together. Now, some people say an agency or AI agent group chat between agents is just a gimmick, but really, if you've got two agents working together and you make it really easy for them to collaborate and talk and debate between each other, that's super powerful. Then Claude actually came up with a cinematic image prompt for Hermes studio, which you can see over here. Now, why do we do that?

Because when we're creating this idea, which we've got over here for this app that we're building out, basically, Claude can't create images. So how do we get better images created using a different agent? We can hand that over to Hermes, and we've built our studio over here where it can generate videos, it can generate voice notes. And it actually created this image, as you can see, with a bunch of others, but this is the image it created for the idea from Claude.

Perfect. And then it actually built out the game and the idea that we had. So we've got that over here, you can see it, like so. And so they can work really well as a team together.

Now, something else that we did is we plugged these into Paperclip and they work together as a team. So let me show you an example of this. If you look at this setup, we've got Claude as CEO, and then it delegates tasks to Hermes agents. And so they can work together as a team to build whatever they want.

Here's another example. We've actually got a full team of AI agents working together. And so if we look at what's been created recently, you can see a bunch of stuff that's been shipped and built today. And then we have multiple different projects, right?

So the AI agents can work together as a team to build whatever they want. And if we want to create a new agent, let's say, for example, we can go to the advanced configuration here. We can select, for example, Claude code. We can give it a name, like, for example, SEO agent.

Hit create. And now we've created an AI agent that's working with our team of other agents. So if we look at the org chart now, we've got the SEO agent. And so now we just need to make sure that we configure it to work under a team.

How do we do that? And so what we're gonna do is assign that to a manager. And so you can have teams of agents working out together. And that's basically what we're doing to create all of this stuff here.

So we've created multiple different projects and multiple different websites with a team of agents working together. So you can have Hermes, you can have Claude, you can have whatever you want working together as a team. So you can see the group chat here, which is one use case. You've got the pipeline of ideas, which you get from the actual group chat, and they add it automatically to this pipeline.

Then you can actually approve the ideas here. And once you've approved them, you can see what was built previously. So if we scroll down here, you can see all these different projects we've built. And actually, if we go over to the gallery, we can preview everything in one place, which is pretty cool.

So we can see all this cool stuff that we've built. And we can open up full screen and look back at other projects. Now, the other way that we can get Hermes and Claude working together is basically by having a Obsidian second brain. And I actually call it the memory galaxy because each memory here is a star, and each star is a different memory and different log about me.

So you can see, for example, here, it notes information about what sort of MCPs we've used, or what content we've created recently, or what decisions we've made. And so everything is linked together, and our AI agents can work together, pulling in context and coming up with new ideas, which is pretty amazing when you think about it. Now, we also have these recent memories, as you can see. And so if we look at when Claude and Hermes are working together, we've got notes on what they created.

And these are actually stored in the room notes from our AI agents group chat. So just to recap here, we've got a mission control, we've got Claude and Hermes side by side. Then we have the AI agent mastermind, where your agents are working together, and you can actually see previous history and chats that they've had together. And then every time they come up with something good, they add it to the pipeline.

From there, we can build that stuff with our agents. And if we wanna build out like a full organization chart with agents working together in a team, you can see that right here. And so this is probably the best way that I know to get your team and your agents working together as a group so that you can get the best of both worlds. And bear in mind, when you look at it, Claude Fable 5 is amazing, but it's also gonna use twice the amount of tokens that Opus uses.

And it's also only available until the 22nd of June. So you might as well use it as a brain, and then get Hermes as the hands to actually get stuff done and implement it. And when they work together as a team, you get a really powerful combination. So here's another example with Paperclip, where you have Hermes as the hands getting stuff done, and Claude as a brain.

So for example, Ada, the CEO today, wrote a launch memo and set three priorities. And then that gets implemented by the rest of the team of Hermes agents. And we can see everything that was created today, and recently, as you can see here, right? With different landing pages.

And then they actually take notes on what they've built, and what they've created, and what they've added, which is pretty amazing in itself. So you can see everything that was built, and these guys can just work together autonomously. I don't need to manage them, and I can just see the work and check it later. And if we have any comments, we can leave a comment here and just say, hey, we need to change this, or we need to redesign this, or great work, whatever.

And then also, what Hermes is really good at is filling in the gaps that Claude can't. So for example, Hermes can create images that Claude has ideas of. It can generate videos. It can generate voice notes, right?

And so, for example, we've actually got this section here, where Claude could come up with an idea, and then from there, Hermes can generate the actual concept. And so they both work together as a team. Now you might say, okay, why not do different things? Like Hermes is great at getting stuff done.

It's great at organizing stuff on a schedule. It's actually great at building out a team of sub-agents. But then Claude Five is really good at thinking coding. So every agent and every model has its own strengths and weaknesses.

Now, let's talk about the old way versus the new way. So the old way, for example, is like you have Claude inside one terminal tab, and then you have Hermes inside another terminal tab. And you're just like copying and pasting between two or trying to organize them both. That's not very useful.

Instead, what you want is this agent operating system that we have. You can build this out yourself. You can create your own system. And the great thing is, if another agent comes out tomorrow or something better than Claude Five comes out, no problem.

We can just add it in and plug it in. You'd never feel behind. And so the old way is like you have Claude in one tab, Hermes in another, and you in the middle. The old way would be copying Claude's plan and pasting it into Hermes and then copying back.

And also the other issue is that neither AI knows what the other one said. And so ideas just get lost because nothing gets built. And there's also no shared memory between what you've shipped as well. And the end result with that is like you have two different subscriptions and you end up very tired as the messenger between both.

With this process, you have both agents in one room talking to each other. Hermes pitches, Claude judges against your real build history and obsidian build history. Then you have Claude's prompt that can be plugged in, for example, to Hermes Studio. And you can generate concepts and ideas and videos and stuff like that.

And then also you can have Paperclip running with Claude as a brain and then Hermes sub-agents working underneath it. And the result is that you have a design studio, you have a dev team, you have a full AI agent company working as a team together. And you might say, isn't this going to use up a lot of tokens? That's the whole point.

It's like Claude is just a brain, so you use minimal tokens for the brain. And then Hermes can run on free models, for example, like local models, or for example, Nemetron 3 Ultra is actually free through Nose Portal. And so you actually save tokens doing this. So just to recap in terms of what you learned, you stopped being the courier and Claude and Hermes can talk directly to each other.

You got a creative director, so Hermes generates and Claude can actually judge and analyze the stuff Hermes comes up with. You got a design studio where Claude comes up with prompts and then Hermes can create it. You actually got a dev team because the ideas you get from Hermes can then be plugged into Paperclip. And you've got these teams of autonomous agents working together.

And also you can see everything. This is the other benefit, I can see everything that was built over here. I can check out the gallery and see what was created. I can also go to Claude and go to the workspace and see what was built here, as you can see.

So this is just some cool stuff that we built with Claude 5 earlier today. And then also we can see what was built with Hermes because we have the studio, we have the video, the voice, the images, we have the previous sessions and the workspace. And we also have an obsidian log where everything we create inside the agent operating system is plugged into our memory galaxy so we can come back to it later and get the most out of it. So that is how to use Claude 5 with Hermes as a team.

If you actually want these working for you, not just impressing you in separate tabs, grab the agent operating system inside the Air Profitable Automate. It turns Claude, Hermes, OpenCore into one system with shared memory, shared context, and one dashboard you control. And you also get the zip file inside there so you can install it quickly. Every prompt, the obsidian memory setup, coaching calls where we can wire it together step by step.

And there's 3,500 members in there, daily tutorials, a 30 day roadmap, and a member map to find builders near you. If you wanna get that, link in the comments description or go to theairprofitable.com. And then if you go over to the classroom here and then go to the agent OS section, you can see the video tutorial and we just updated it today so you can see the last update date. And you also get the zip file over here.

And then you also get new daily tutorials. So we've actually done a full masterclass on Claude 5, how to use it, the best prompts, how to implement it, a 30 day roadmap. And every time we learn something new that's actually useful about AI agents, we create a full video tutorial, a step-by-step guide, and a roadmap on how to implement it. And then you can post questions inside the community, you can jump on weekly coaching calls, and this is all inside the AI Profit Boarding.

So thanks for watching. Hope to see you inside there. Cheers, bye bye.