Hermes + Obsidian Just Changed AI Agents Forever

Hermes, Obsidian and OMI equals the perfect memory system. I'm going to show you exactly how today. So right now your Hermes agent wakes up every single day with no memory of you. It doesn't know your business.

It doesn't know your goals. It has a little bit of context, but it's not very good. It doesn't know what you really worked on and it doesn't have full context. You kind of have to explain almost everything from scratch every single time you open it because the memory is not powerful without what I'm about to show you.

Now I'm about to show you how to fix that for free with three free tools, Hermes agent, Obsidian and OMI. You're going to learn how to wire them together and your Hermes agent will build a perfect memory of you automatically, your goals, your projects, your schedule, your preferences, and it gets smarter about you every single day without you lifting a finger. I'll show you the exact setup, why it works and the specific prompts I use to get Hermes to come up with ideas and automations I never would have thought of myself. By the end of this video, your Hermes agent is going to know you better than most people you work with.

Let's get into it today. I'm going to show you how to make your Hermes agent 100 times more powerful using the powerful memory system that I've set up that's super easy to set up and really useful, right? And this is using something called Obsidian, which basically stores all of my memories and all the stuff that I'm working on day to day and connects to something called OMI. Now if we have a look at this side by side, basically inside Hermes agent and any agent that I actually use, we actually have this connected to all of my notes, right?

And so the cool thing about this is we have OMI, which actually tracks what I'm working on day to day. It actually records my screen and records my microphone, which is pretty crazy, but super useful. And then it takes that information and it creates notes day to day. All right.

So you can see, for example, here is actually taking all of my latest notes and interesting stuff. And you can see when it was last updated and it has 943 memories written automatically about me. Right? So for example, here, what this is doing is inside OMI, we are recording everything, right?

Everything that work on day to day, all the work that I do, et cetera. Right? Then we sync this with something called Obsidian. Now OMI is free to try, especially for the features I'm showing you today.

And then Obsidian is free to use as well. Right? And all this does is it takes everything that I'm working on, condenses it into a super simple framework, like little notes, little bullet points about me. So it's like a note taker, taking notes about me all day or like a little assistant, just gathering loads of information about me.

And then it plugs it into Obsidian. Right? Now what we do is we go to our app section inside OMI. We click on sync.

Now we export that to Obsidian vault and this updates all of our memories. So now we have 1,091 memories about us. Right? I didn't write any of these.

It's literally just taking notes on me all day and then configuring my memories inside Obsidian to make sure that we've got everything here. Now, there's a few ways you can use this directly. Even if you don't use Hermes agent, it's super useful because I can actually have a conversation. I've got my tasks.

I've got my goals. I have a little reminders here to stay focused, which actually pop up whilst I'm working inside OMI and we can use OMI like chat GPT, right? We can speak to it right here and ask it any questions inside the chat. Now, also it's the same inside Obsidian, right?

So inside Obsidian have all my memories. I can see when it was last updated and we have this OMI memory export file, which is just a markdown file. And then what we can do over here is reference that particular chat over here and we can ask it anything about me, right? So we can actually use this to be like, Oh, what was that thing that I was interested in or what was I working on last week, et cetera, right?

So you can speak with the AI inside OMI. You can speak with it inside Obsidian. You have all of your memories and your context and everything about you basically stored inside here. You can also organize this.

You can get the AI to organize all your information, but I just use a long list of memories like you can see with OMI here. And then here's the cool part. We can go inside Hermes agent, right? Inside a terminal here and we have our Obsidian vault and our memories configured with Hermes agent, right?

So you can read the latest memories and then it, and then it can use that to generate better use cases and understand me far better. Right? So if I go inside here and I'm like, okay, based on my memories vault with OMI, what are some things that I could improve? What are some things I could automate?

What could you help me with specifically? And so this just helps you get far more out of your Hermes agent because it's way more personalized to you. It has better context on you. And also everyone's, I love Hermes, but I don't know how to use it.

I don't know what the use cases are. Right? And so if it has access to all of your memories and all of your details and full context of you and everything else, then it can use that to give you better ideas on how you can work with it. Right.

And it can be proactive in terms of helping you. So you can see, for example, here is, here's what I could do, right? I could actually take your memories and turn them into dashboards. Right.

And then I can use that to create like a dashboard for this project. And that project could be a dashboard for SEO. It could be a dashboard, for example, like improving retention in the business, et cetera. Right.

Then also it's giving me, given some ideas on like how it can reduce distractions. Right. So as she says, your OMI log shows a pretty clear pattern. You're distracted sometimes, and here's how I'm going to help it.

Right. The issue isn't that you're lazy, but it's, here's some things that we're going to fix. Right. And then also it's, here's what I think you should automate, which is pretty crazy.

Right. And so it's got so much information on me, what I do, what I work on this, here's what I think you should automate, right? You should have a daily OMI digest every day, scan new memories extract like the top restrictions, team improvements, content ideas, et cetera, and then write a short daily brief inside Obsidian. Right.

And so that is something I've never even thought about doing, but it's, this is probably the best thing you could do right now. Right. And it's actually, we could do a memory cleanup, so we can improve the memories and that sort of thing. We could come up with a content idea extractor, right?

It's like you've got loads of scattered content signals inside your memories. Here's how we could extract that and organize that properly. Right. And here's the automation that I recommend.

And so Hermes agent just becomes like way more powerful when you're using this, because it understands you better than you understand yourself. It understands you way easier and it comes up with interesting stuff. It connects the dots in ways that you'd never thought of before. So you can use it for coming up with automation ideas.

You can use it for coming up with systems that you could build. And you could also use it for coming up with like even weekly performance reports. Right. So every week, for example, we could answer like, what do you spend your time on?

Why she moved the needle in some ways, what should be delegated, what things are repeating, what bottlenecks keep showing up. Right. And it could create that report directly for you. So many cool things that you can do with this.

And the thing that I would say here is like the most powerful AI agent isn't the smartest one. It's the one that knows you the best. Right. So when you're using Hermes, if it knows you better, you're going to have way smarter agents, right?

Way more powerful agents. So this is a free system that you can use that gives your AI agent a perfect memory, right? It means your AI knows your goals, your schedule, your preferences, your business. Every single time you open it, you don't need to repeat yourself.

You don't need to start from scratch. You don't need to waste any time explaining who you are. Right. And this is exactly what I'm using right now.

As you've already seen. Now here's the problem as well is imagine you hired an assistant, but every single morning that system wakes up and barely remembers anything about you might have a few rough notes, but Hermes is not going to know everything about you. Right. So you have to explain everything over and over again, your name, your business, your goals, your schedule, et cetera, every single day.

Right. That is basically what using AI agents is like without a memory system. It's pretty exhausting. It slows you down, waste hours, right?

And this basically solves it. So there's three tools that you need. All of them are free. Number one is Hermes agent.

This is your AI agent. If you've never used it before, you can get it on GitHub at Hermes agent. It's just a super smart system that lives on your computer, right? It can do tasks for you.

It can ask questions, it can run automations, it can help you work faster. The problem is by default, it doesn't remember who you are between sessions, not in the way that you want it to. You'll have some rough notes, but it's going to be nowhere near as useful as it could be. Then you've got tool number two, which is Omi.

So Omi, you can even combine it with your wearables and that sort of thing. This is another open source project, right? So it's a free app that runs in the background of your computer, listens for your microphone, watch your screen. Obviously don't give it access to anything you're not comfortable with.

I just allow access to that because I like to test things and it quietly takes notes on everything you do, you say, and you work on all day long. And then it organizes those notes into clean, readable files, right? And memories literally knows what time you go to sleep or what projects you're working on. It knows what music you like.

It knows your goals, right? So that is tool number two, which is Omi. And then the final tool you want is Obsidian. So Obsidian is like a free note taking app.

It's like a filing cabinet that lives on your computer. And Omi basically exports all of your memories into this filing cabinet automatically. And because Hermes agent can read all your local files, it can read your memory file too, which means your AI agent wakes up every day, knowing everything about you and knowing more and more about you every single day, which is super powerful, right? And you can connect this to open core.

You can connect this to any agent. It doesn't have to be Hermes, but basically the big idea here is like Omi records your life. Obsidian stores your memories, Hermes agent reads them, and your AI finally knows you. That is the biggest improvement here, right?

Now let's talk about some limiting beliefs I see holding a lot of people back with this sort of stuff, right? So number one, people say AI memory tools are complicated and technical. Honestly, if you can click a button and drag a folder, you can do this, right? Omi installs in minutes.

Obsidian is a simple download. The Hermes connection is one line of text telling you where your files are. So there's actually no coding. There's no technical setup.

There's no experience needed, right? Other people say I don't want to create a memory vault for my agents because I have to keep updating the memory manually every single day, right? Really the system updates itself automatically. You literally do nothing.

So Omi runs in the background. It captures memories whilst you work, exports to Obsidian without you lifting a finger. Hermes reads it automatically when you open it and there's zero manual work after the first setup. And the setup is literally download Omi, download Obsidian, go to apps and sync.

That's it. That's literally as easy as it gets. And then other people say, I don't have the resources to do this, but actually the entire system is free, right? Omi is free.

Obsidian is free. Hermes agent is free. Omi does have a tier on it, but you won't be using it for the chat. The only tier that is on this is if you're using the chat, which is here, but you're not using this, right?

You're using the sync to Obsidian section, which is free. So Omi is free. Obsidian is free. Hermes agent is free.

And that's basically it. Other people say my AI doesn't need memory. I can just explain things each time. Honestly the right belief here is every minute you spend re-explaining yourself is a minute that you could spend doing something way more productive, right?

Think about how many times a week you repeat your context to the AI. Now multiply that by 52 weeks. That's dozens of hours lost every year and they're just gone, right? Whereas a memory system that literally saves you time from the very first day that you start using it.

Other people say this only works with Hermes agent or whatever tools, but actually that same memory file can work with open cloud code and any AI that can read markdown files, which is literally every AI, right? So Obsidian vault is just a folder of text files and any AI that can read a file path can use your memories, right? So you have one set up with every tool. So that's basically it.

We actually have a full framework on how to use this, which I call the Julian Goldie memory stack framework, right? The infinite context engine. This is basically a new paradigm for AI agent workflows built for people who want the AI to work like a super powerful assistant, not a forgetful intern, right? So most people are using AI agents the wrong way, right?

The other way. They open a tool, they type a long paragraph explaining the VR, they get an answer, they close it. Tomorrow they do it all over again, right? That is the blank slate problem.

Your AI is super powerful, but it's like a genius with amnesia, right? Every conversation you're starting from basically zero, whereas the infinite context engine breaks this cycle permanently, right? And so you capture everything with OMI. You organize your recordings into clean structured memory notes using Obsidian.

You store that inside the memory vault, and then you deploy that inside your AI agents, right? And we actually have a full guide, SOP, 30 day roadmap, and a hundred prompts that you can use for doing this inside the AI profit border, right? So if you go to the daily update section here, you will find a full guide on the infinite context engine. And basically we walk you through the tools, the setup, the limiting beliefs, the full framework that I talked about today, how to implement it with step by step operating procedure, and also a 30 day roadmap to really get this running inside your business.

Plus a hundred prompts that you can use based on your memory, right? So for example, based on my memory, what content topics could I use to help me grow? Or what is my brand tone of voice based on my memory file, right? We have a hundred prompts like that to help you really get the most out of the system.

So feel free to get that inside the AI profit boarding link in the comments, description, or go to the air profitable.com. We actually have 158 pages of people winning, learning, and growing with all of these different tutorials that we're sharing inside there. So it's just an awesome community of 3000 business owners getting the most out of AI automation, learning, growing every single day, and just really connecting together as a community on this journey towards learning and scaling with AI automation. So if you want to join and be part of that, feel free to check it out inside the community.

You can ask any questions, get help and support, et cetera. Inside the classroom, you get access to all of my new daily tutorials, including my best trainings on OMI, Hermes agent in general. We've got video tutorials, step-by-step guides, et cetera. And then we also have four weekly coaching calls.

So you can jump on these life coaching calls, share your screen, ask questions about your setup, connect with other people doing similar things to you. And also inside the map, you can connect with people in your local city who are also using AI agents and AI automation, just like you. Plus we have tons of training inside here on Hermes, on OpenCore and AI agents in general, plus AI automations. Feel free to check out link in the comments, description or just go to the air profitable.com.

Let's see what questions we've got here. So Hermes would be better for more creative projects. Yeah. It's really good for that.

Like any AI is pretty good for creative projects, but for sure you could use it for stuff like that too. Claude Mem says, I want to say I'm going to totally use Aon UI. Been in my plans. Yeah, it's really cool.

And thank you for covering Claude Mem before. Happy to help mate. Thank you very much for commenting. Well done on the amazing project, by the way.

Do you use any Microsoft or Windows co-pilot, AI and productivity apps? No, I don't. I just use a Google drive. I don't tend to use anything like Microsoft word related or anything like that.

I appreciate you. Happy to help. I saw that it's possible to integrate Higgs field on Claude's website. Any alternative?

Yeah. You could integrate Higgs field is like a MCP, right? So you can use that with any AI. Like you can connect that to Hermes or anything like that.

It doesn't have to be Claude. Claude is just really good at connecting to external tools and that's why most people use it. Hermes desktop apps. Which one is my favorite?

Yeah. This is my favorite right now that I've tested out Hermes desktop is called. It's really powerful. Like it comes with the office chat.

You've got multiple profiles. You can manage tools, settings, skills, et cetera. It's called Hermes desktop on GitHub. That's my favorite app for managing Hermes on desktop.

Can you have multiple agents under a single profile? No. So each profile is an agent, right? So you wouldn't like, you can give it multiple skills and give it multiple powers, but every profile inside Hermes is a separate agent.

If you have the same agent, then it would be the same agent every time. Right. And so it wouldn't be that multiple agents, but bear in mind, like it's really easy to create multiple agents inside Hermes.