Obsidian 1 HOUR COURSE (2nd Brain for AI)

But it's one hour free course on Obsidian. I'm going to show you exactly how to give your AI agents a second brain and how to really take advantage of Obsidian to make it the most powerful context engine you've ever created. And this is a really powerful system, whether you're using AI or not, because you basically have a whole database of everything in your life. Get AI agents to analyze it for you.

And it's very, very powerful. It's literally changed everything for me since I've set it up. So in this one hour free course, I'm going to show you how to use Obsidian, how to get the most out of it, and how to build and automate anything with your AI agents, plus how to get your agents to actually automate and set this up for you so that you don't need to do any of this yourself. Now, what is Obsidian?

Obsidian is simply a note taking app. So you basically have a way of storing all of your notes in one place so that you can come back to it whenever you want. So let's have a look here, for example, if we go over to our notes and we're checking out everything that's created over here, you can see that we have notes and files for everything inside our database. And so what this means essentially is that we can easily have a full system of context and we have like an organized system where everything in our life is detailed into neat and tidy documents.

So all of it together inside the knowledge. Now, one of the benefits of using Obsidian is that if you're switching between AI agents, so if you've got Claude, you've got Hermes, you've got Antigravity, you've got Codex, you've got GLM 5.2, et cetera. Well, how do they communicate with each other? There's really only one way to get them to communicate with each other properly.

And that is with the Obsidian database. There are other things you can use that are a competitor to something like Obsidian. So for example, you have something like Graphify, but Graphify is just for like one single code base. A lot of people ask me, why don't you set up, for example, Graphify instead of Obsidian for something like this.

But Graphify really, it's just like for looking at one single code base. Like for example, visualizing one GitHub. This whole system right here is really for creating a theory for the AI agents. If you don't feel how to get access to it, you can go over to Obsidian MD, obsidian.md, and obsidian.md is a free app that helps you download Obsidian.

So you can just click on get Obsidian for macOS. And once you've done that, you can set this bad boy up. And so it's just a really good way to have AI agents. The other thing to see is there's nothing really that special about Obsidian itself.

Like we're not like reinventing the wheel here. This is literally just markdown files and folders. It's a full wiki of everything inside our database. So you've got this base, for example, it's like, you're going to look at this.

Obsidian is how it's fastened. It looks a little bit like a memory galaxy, which we're looking at right here. It looks amazing when you go inside the knowledge graph as well. But don't let that fool you.

These are just files. It's a very boring system. It's basically folders and files stored as something called markdown files. And the markdown just is, the actual content itself is written in the markdown.

So you see how there's, for example, a hashtag next to NotebookLM, that means that it's a heading. So one hashtag equals heading. Two hashtags equals a subheading. And then you can see, you can format it with bullet points and everything else.

And so what you'll see inside our system is that we have, for example, a note on the grade here, we have the properties of it. So tags, it's all together. And I'll show you some interesting ways you can organize that later. The maker.

So this is specific to our database. Basically we've created a section here where it details any tools that we use. And then the maker is, what category that falls into, what status that comes into as well. And then when we actually look at this one single markdown file inside a whole database.

So it's just a Wikipedia. But it's Wikipedia for your brain, for your memory and everything else. And you can see where it connects to. So for example, Google, Gemini Antigravity, Julian Gold SEO, et cetera.

All right. So you can see, for example, we can click on this and it'll take us to our title formula. This is a title that we'll typically use for our content. It works really well based on what's worked in the past.

And here by the way, it's right here. Now, you know what you're thinking of doing, Julian, this is great, but this must take a long time to set up, to write out, to organize, to think about. The beautiful thing about this is that actually when you're doing this whole system, you can get your AI agents to update it for you. And they can basically look at what you're doing day to day.

And then from there, they can actually create the whole process. And that's how we've done this. So if you look at these notes on the left-hand side, I've filled a key page up here. This is all organized into, for example, like the archive, the memories, the areas, projects, daily notes as well, daily logs, and then we've got like little subfolders, like the Air Profit Boarding sales page, or for example, our conversion playbook and everything else.

So you're doing this, you can get our agents to be able to automatically organize it for us, this infinite context engine. So what that means essentially is that we have something called OMI. And OMI is a powerful system that can actually look at my screen, analyze what I'm saying inside the microphone. And it can also look at what I'm working on day to day.

And then it can take notes, and eight days later, they'll be able to see it later. So you can see some examples here of, of recent memories that have been organized in here, so three days ago, four days ago, and we can see the timestamps of everything that we've done recently. This is really good because I don't have to organize this myself. My team can, or my AI agents can.

So let's talk about that in a second. So this is the infinite context engine. Basically it's one memory that every agent shares and you just have to train your vault, your vault trains for agents. It's a loop that gets smarter forever.

Now, if you wanted to just get started with something really basic on OBS-EDA, you can download it. So you go to OBS-EDA, download it for free, and you can get it on Mac and Windows as well, and once you've downloaded it, you're going to be able to use it in your And once you've downloaded it, then you can go over to something like Claude and what you can actually do is use a prompt like this, so you could say to Claude, okay, we're going to take our local directory for OBS-EDA and so if we click on open folder here, and then we go to our documents, we can find any folder to use OBS-EDA like this, so you've got OBS-EDA there, stored inside our documents. And then we can open that folder. And that is basically everything that's inside this app, inside the Obsidian app, is now open inside Claude.

And you can do this with any AI agent. You can do this with, for example, Claude or Hermes or Opecor or even Chachibity, whatever you prefer. And if they go inside Claude, you'd say, okay, create a list of recent projects you've worked on, and then connect my Obsidian vault inside the folder and add a new file called latest projects. And I just want to show you really simply how you can create a basic markdown file inside that list and be really detailed in terms of everything you've worked on lately, how you've helped, what you've built, et cetera.

So this is how you can create your first note inside your Obsidian vault. That's how you can do all of this. And so it's going to begin working on that right now. And you can see that it's beginning to work.

It's magic in the background. I'm actually, good little tip for you. If you're using Claude desktop, like I am, you can just pin the thread and then you drag it to the top. And that way you can see the latest stuff, as you can see right here.

So the detail in this conversation, we can see everything that worked on and we can go back and forth with it. This is actually looking at recent memories inside a vault, and then it's going to come back to us. Another option that you have, and this is a lot more time consuming that I don't personally recommend it, but you could just create a new note here. And then you can put in the details of.

No, okay. This looks great. So I know how to create a markdown file. And by the way, you can add like bullet points here.

You could add thoughts. You could add subheadings like so, but again, I don't recommend doing that. I recommend using your AI just to do it for you. It's going to save you time.

It's going to initiate to use. And it's also going to make sure that your AI agents are responsible for so that you delegate managing the second brain in the long run to your agents. So just to recap on what we've learned so far, we've learned what Obsidian is, why it's important in terms of giving a memory to all our AI agents, how to get Claude to organize it for you, or any AI agent, I mean, you could do the same thing, for example, the Appendix is where you type the Appendix in. And you're going to get to, okay.

Create a new memory inside my Obsidian vault locally. And we can ask it to start doing that for us as well. And that's another option as well. And when you're using AI agents, because they work with markdown files.

So for example, if you have a SKU file, you're the AI agent, and the markdown file is built on your SKU. to get it to stick. So they're doing this to understand it, what they are. Now you can see it's created this latest project section on the vault room.

And so if we open that, as you can see here, we can open it in finder. And then we can use an app to open this up. So I'm actually just going to check inside Obsidiate. If search, if you search inside Obsidiate, you're gonna option it.

And then you can type in latest projects and we can find the file that's just created. And as you can see, if we search here, to find a note in the map, I guess it's a file that it just created for us. And it's way more organized than I am. It understands how to write better markdown than me and it can create all these notes.

So you can see here, it says updated by, and it says code Fable 5. Here's the date that it was created. Here's the tags. And these tags are just ways of organizing everything that you're working on.

So this is what, even if it's different markdown files, so you have your, so don't, don't go to the internet for it. The only way to go to the internet for it to make them connected is with tags. And tags are important because when your AI is looking at the notes and trying to fill context for you, you really need to have a system where it can easily see all the connections and all the links between everything else. So one of the reasons that we created this memory galaxy is basically to visualize how our AI agents see our Obsidian database.

The better we can do that, the easier it's going to be for us to make it more agent-friendly. And the more agent-friendly it is, the easier it's going to be to use day-to-day and the more powerful it becomes. So this is the whole system here. Now, obviously we have thousands of menus here.

You don't need that many. Like, even if you just get started with like three or five a day, you can do it in that day-to-day. This system improves every single day that you use it, essentially. Now, if we have a look here, it's like, right, here's the latest projects you've worked on.

Here's a detailed log of what you worked on across all of these different projects and actually knows everything about it. So we've taken the knowledge from Claude and then we've actually said to Claude, okay, organize it into a Mardum file. And so you can see for example, it's looked at the shared brain system. There is a bonus that this agent that we built into our agent S recently, HY3 and actually dates it as well.

So we've got the date of when we worked on that project and we've gone with that. Now, at this point as well, you might be saying, okay, why don't you just use Claude? Why don't you use Claude for your memory? Why do you have to have a separate second brain?

But the thing is, Claude doesn't link naturally to something like Hose. So if you use, for example, Claude and go after Hose, it just doesn't have any context to what you worked on day-to-day. And so it's not linked together and that's what we want. We want everything linked together so it's easy to retrieve the data.

And bear in mind, like all of these AI agents, they come and go all the time. So for example, if you're looking at something like Fable 5, well, that could take you down and then it came back out eventually. But the thing to note there is like, you don't need a system because you can't control the model. And you can't control what the model does.

And so you might as well have systems that you improve every single day and then it doesn't matter which models come and go. So if we scroll through here as well, the other reason that I would get the AI to organize it and not you is because there's a lot of boring admin work here that would literally take you days and days and days of work, if not weeks. As if you get your AI agents to do it for you, they save so much time and you don't have to be extremely bored trying to organize it yourself. The other thing that I would say here is like, quite often, you know, if you're using something like Claude Code, it's not going to take shortcuts in organizing the whole system.

Whereas for example, if you were and you had to add every single little tag and every day and who it's updated by and like, it'll just either do all that weird stuff. If you're doing it, it'll take years and years and years of work, right? You're like, trying to do it to the extent that you can of your mind. That would be extremely time consuming and boring.

So instead, we'll build this out. Now, also, if we have a look at this system, one thing you're going to notice is that it's all linked, right, so it's all neatly color coded. And the things that are big, as you can see, so for example, this big orange dots, these are the projects. But then you have smaller dots and these are like different tools that we use.

So for example, we have over here, details on a maker index, which is a tool we use for SEO. And we can see what that links to. So the interesting thing about this is we can see how everything connects together. And that's really what your life is about, right?

It's like one thing leads to another, you try out more than actually leads to homies. Try out homies, that is the open port. It's really intended in this interesting way for the information that you can use to basically help you understand your world. I also think it's very interesting to just see how your world is organized.

Now, you might be thinking at this point, okay, well, we know how to use Obsidian. We know why to use it. We know why it's more powerful than using the default memory inside any AI tool. We know how to get called to create a learning system beyond that file.

At this point, you might be wondering, okay, well, how'd you jump from getting that single markdown file to having a full organized database? So what we're doing from here is we're gonna say, okay, based on what I do day to day, how I work, what I use AI for, the projects you're gonna go, does it all work against the different AI agents? That's a really good prompt to extract all the memories from each tool we use. And then also, how do you recommend organizing the Obsidian database so it looks beautiful and it's very organized?

So two different tasks that we need to do there. Number one, we need to get a good prompt for extracting the memory that we already have inside our different AI tools. And the other two, we need a system for organizing it. Now, you've actually used something called Para with this.

So the system that we're using inside this whole thing is called Para. If you're not sure what Para is for project management, let me call this up for you and you can have a look and see exactly how it works. So basically, it's a simple way. It's a universal organization system.

The structure to a digital life, it has 30 categories. So projects, and that could be short or long-term, goal-oriented tasks, areas. So long-term responsibilities, resources, reference materials, and archives, which is the completed work. And that's what I've used to organize this system.

Now, you don't necessarily have to go with that. You can pick your own. You can make it your own. I do think it's very important.

You know, you do it in any sort of system to be able to look at things like that, right? And instead of just copying and pasting it up for the app, because that's gonna make it way more, you know, there might be things that you see here. You're like, I don't really need that. I don't really like that, et cetera.

So if you go back into Cloud Now, it'll actually give us some ideas on how we could organize this. So you can see, for example, what it's actually done is given us a prompt here. As you can see, now it's actually put that page inside the AI definitive, and it basically gives us a way to extract all the existing memories. Because if you're building this from scratch and you've been using Chats, UBT, and then Cloud, and then Hermes, well, this helps you explore the whole memory and create that memory.

So you've got a starting point. And then after this, I'll show you a way to get the AI agents to automatically update it for you. So for example, if you copy this page over here, and if you're using something like Chats, UBT, day to day, then we can actually paste that prompt into Chats, UBT, and you can use it inside any single tool that you use. But the whole point is that you can easily extract everything that the AI agents know about you.

And then it can give you the full details here. Now you'll see as well that it says, write this to my obsidian board. Obviously, if you're using Chats, UBT in the cloud, it's not going to write it for you, but you can just delete that step. If you want to get the full prompt from me, you can get that inside the classroom.

What I'll actually do is I'll add it inside the new daily update section here, and we're going to have a full step-by-step notes set up for all of the course from today. But then also, if you want any of the prompts that make sure the AI fails, you can go to today, update it here. So this is a prompt for extracting the memories from every agent you use. And if you're looking for the link to obsidian, it's just obsidian MD.

I'll link that at the top as well. All right. And we're going to start with the basics here because it's important to do these. And then what you can do later, you can do the authenticated stuff.

But the start is get obsidian, get your AI agents to give you the memories, and then you can go from there. So if we have a look, for example, here, it's beginning to extract the memories. And we can do this inside Hermes as well. So if we go into terminal here, and we're actually going to switch over to Hermes model.

And if you switch to a different model, let's go with, let's say, we're going to go here. I'm going to do chat2bt and open AI, just to make it easy mode for us. So we'll go with codecs, and then we'll use the existing examples, GPT 5.6. And then we're going to start Hermes like so.

And then we're going to grab this prompt. What we're doing is basically getting all of our agents to dump their memories, give us everything that they understand about us, so that we've got this really powerful starting point. And the great thing is about this as well, is that you don't have to start from zero. you're gonna save time, you can get your AI just to just do most of the heavy lifting for you.

So that's the plan 2.7. And then we can give this per tool. So we could use Insight Grok, Hermes, Codex, Cloud Code, Kimi, whatever tools you're using day to day. Now, at this point, I'm gonna ask actually, what are some of the best ways we can set up Obsidians?

For example, I use Para. Other than any other, it's just structuring the whole project so that we can get more out of it. And also, another really powerful benefit of this is like, it's gonna understand everything about who you work with, which is great for delegating tasks. And then also, it's gonna understand what projects you work on.

So for example, we've got the details on the AI Profitable Room over here, as you can see. It's taken all the notes, it's added the tags here. And then inside the setup, it's all interlinked together and it's all neat and tidy. So it's got details on the funnel, open threads, etc, everything else and where it connects to.

And so when it understands your projects like this, if you ask for ideas on what you can automate, that's gonna save a lot of time as well. Now, let's have a look at some ideas. So you get Para, which is this bit here. And then you also have Zettelkasten, which is like, basically, kind of like one idea each per atomic note.

Lots of heavy linking, but it's very flat, there's no folders. And this is interesting. This is quite a lot of friction to set up, a lot of discipline, and probably you're yelling for the speed that I work at. But if you're interested in doing something, be able to do it that much quicker, just like in your own project, then you can actually have a flat folder.

For me, I have too many tools, too many people, too many projects working on these flat setup like Para. So I would tease a flat setup like Zettelkasten. So I would stick with Para personally. It's got maps of content.

This is not something I'm familiar with. But basically, everything just links to each other, it looks like as well. So there's all sorts of beautiful ways to organize idea notes. But I'm gonna stick with Para.

I was hoping it would give me some ideas on something better that I could use, but actually, no. So we've got this whole system. Now, when you're setting this up, it didn't, for me personally, it didn't look like a beautiful round circle like this. It was kind of like all over the place, and it was very, very messy.

And by the way, I should take you through at this point, what each section of this actually is. So when you're using Obsidian, you've got all these tools on the left hand side. So we've got the quick switcher, which means that we can quickly find or create a note. So some of our most recent notes are over here, we can just click on them and then go straight into them.

Then we have the knowledge graph. And this just basically shows you everything and how it is related to each other inside of the notes you have. So if we have a look at this, you can click on the settings over here. And you can actually, for example, add even more detail to it.

So you can see when I add the tags, this whole structure gets way more complex. You can also, if you're on a Mac, you can zoom in and zoom out. So for example, I'm just using my mouse to see it. So I'm actually going to remove the tags.

I don't think it looks as interesting when it's like that. And then also we can remove the existing files. We can also add attachments as well. So if we have stuff attached to each of our files, we can look at it like this.

But it's just a really interesting way to see how your whole database is organized. So you think it does look great. What you can actually do, this is how I got it into like a nice little circle here, is you can just go back into Claude and you can tell Claude, hey, can you just organize my Obsidian knowledge graph? So it looks like a beautiful circle.

Also, it looks more beautiful. It's structured into something that looks way nicer. And that way you can just like make it more compact, make it more interesting, you know, unify it up. You want an Obsidian, I mean, if you've seen any of the videos and stuff, then these sort of things, you'll see that people have like all sorts of beautiful ways of visualizing the Obsidian folder.

And that usually gets more attention than how to actually use it. It's like, usually people are just excited by the way it looks and how cool it looks. Now, what you also see over here is we have groups. So we have people, we have tools, we have companies, we have concepts, places, communities, sites, everything else.

Sticking with people, tools, companies, concepts, very, very important to have those in there. Because again, they are weaved into your world. They are weaved into your second brain. And you can use this for business or for personal.

For me, I actually have a bit of both inside there. So you can see, for example, if you actually had a line earlier, that's actually sort of a part of the treasure isn't it? Just having a nice system like that helps a lot. And then you can also change like the node size over here.

So you can make things smaller, bigger, and you can also change like the text fade threshold. So if you want, you can visualize all the text and what each name is set up as. Or you can just remove that completely, like so. Actually, it looks quite nice when you can see the layers that have been created.

And then you can change the line thickness. Again, these are just like aesthetic changes that you can make to make it even more beautiful. And you can animate it over here. You can change the center force.

You can change the repel force. I mean, honestly, you could spend all day messing around with those buttons. It's not going to make you more productive. So it's a really cute little thing.

You have like forms and a canvas. Don't really look at that too much. We have the daily notes. You can see we can open today's daily notes to see what we're working on.

We also have, for example, the command palette. And this is really good. So if you go to create new base here, you can actually just scroll down and see all the notes at the time. You can see how we do all these different notes.

These are the names of each of the files inside of Obsidian memory. I can see even, for example, and again, I haven't created any of this, but I can see that it's actually taken notes even on my personal trainer who I work with. So stuff like that helps a lot. And then we've got like this whole database, which is pretty amazing.

So that's basically it. In terms of how to organize Obsidian, a rich part is in his face being. There's some other like add-ons to add on the right hand side. I've literally never used this, but you can actually add Claude inside here.

So you can talk with Claude on the right hand side and it's kind of like an extension you can add to Obsidian and then you can ask it questions about your Obsidian database. And you could actually not even have to use Claude desktop. You could just speak to Obsidian directly inside the database over here. For me personally, I knew how to use it.

So that's why I'm not showing you how to set up. I just don't think it's particularly useful. So what I do like though, is inside Claude on Claude desktop, we can just have this conversation for Obsidian. And then anytime we need to come back to it, it's ready to go.

Anytime we need to, for example, ask it a question, it's ready to go. And like, I think that's so much easier because it's got full context at a time, if everything is done inside Obsidian, which is really, really powerful. And you could always fork the conversation if you're working on two things at once inside the full conversation here. So if you go back to Hermes, you can see that it's basically beginning to, it's working now with the setup that we gave you before, right?

So it's coming up with ideas. It's really cool. You know, it's basically what we gave it, right? So we gave it this full prompt on getting memory, it's calling all its tools, and then it's recalling all its memories.

And then it's going to update that inside our Obsidian database. So once you've set up Obsidian, linked it to your AI agents, put in the prompt to get all the memories, what other data do you put in here? You say, organize my Obsidian database so that it uses para. That way, everything is organized in a neatly beautified way.

And then you can just hit enter, and you can get Claude to manage your Obsidian database. Now, I personally think Claude is the best for doing this because it's autonomous, you can just work in the background, particularly the Theta 5, it's very, very powerful. But as you saw, Hermes is a lot slower when you're doing something as big as this. So you can see Hermes is still working on it in the background.

And the same with ChatGPT, like ChatGPT took quite a long time, and sometimes it just doesn't work. By the way, you've got the full export here, as you can see, of all our memories and everything else that we can then just plug into Obsidian. So that is basically how to utilize the Knowledge Graph, and how to get your AI agents to dump their memories into it, and how to organize that big memory dump. And from here, we're going to talk about something called the Infinite Context Engine.

And this is something I've set up, and I think it's probably the most powerful system I've used for having Obsidian fully automated, so that your AI agents, they're on the loop, they basically pull the context out of Obsidian to give you great answers. But then also, they're going to put context into Obsidian on a daily basis to save you as much time as possible. And this means you never have to worry about updating Obsidian again, but it gets better and better and just improves on a daily basis. So let me show you an example of this.

When I go into the preset inside our AGS here, you can see that it is basically updated. in like every hour or every day, we've new daily updates on what it's doing. So if we go into this section, for example, we have all of our memories and every single day, our memories are being updated, right? So we have this decisions log, for example.

It's looking at all the decisions that we've made lately and that sort of thing. Now, the great thing about this sort of system is it's fully automated. It's being updated on a daily basis. You might wonder, how do we do that?

So what I actually do is, for example, for some of the automations inside my AgentOS, I will tell it to log its latest actions, its latest activity into Obsidian on a daily basis. So every time the AgentOS gets updated, it will be plugged into our Obsidian database, and then we can see the history of things as well, which is great as well. So for example, we can see Monday the 13th, Sunday the 12th, we can see all the latest outputs and activities from this, and these are all plugged into our Obsidian database, which is great as well. Also, if you go into, for example, like hoodies, you can say, okay, basically, you know that, right?

Here's some ideas for, for example, for keywords, right? Or that sort of thing. So you can see, for example, here, we've said every day at 6am, recap my to-do list based on what you know about me. And it's pulling that from our Obsidian database.

Now, if you want to use this system, here's the three ways that I use it. So one tool that I use is something called OMI. And OMI is a guide tool that can basically take notes on what you're doing. So it looks at your screen, it listens to what you're saying through the microphone.

It'll be listening to me right now as I talk to this. And what this basically does is it takes me automatically. Now, this is a free tool, you can see. Is there another one here?

No, if you have Windows, then I shouldn't. You probably want to use this GitHub tool instead of OMI, but either way, these are two different great tools you can use here. So this is another option if you're on Windows and you still want to use the system, if you're on a Mac, then you can just go with OMI. The whole point here is it's a secondary.

You trust and actually understand it better than you know yourself, in some ways. So here's what it looks like once you've downloaded it. So you want to download OMI and only give it access to whatever you're comfortable with, right? So for example, if you are not comfortable with it listening to your conversations, no need to set this up, right?

But if you want something that really does understand you, you bet that you understand yourself. This is how you did it. So what you've seen is basically what it did to OMI, instead of the say listening here, and it will understand and look at what you're doing day to day. And then if you go into the memory section, you can see here, for example, even like 18 minutes ago, it's like Julian is focusing on Obsidian, right?

So the user is in Obsidian, here's what he's working on. And it basically tells you, here's what we're doing day to day. It's fantastic. And so it's taking notes of me like every single day.

And then what we can actually do automatically on a day by day basis is we can export that to one of our other systems, right? So step one is you will OMI, that takes some memories. And then if we go to more, inside the connects section, you can actually link this to Obsidian, right? So you see how it says sync, and we've got the option to sync our memories with Obsidian.

We're going to click that now, hit export. And that will now export all of our new memories to our Obsidian vault, which is fantastic. So we've got this whole list of memories that's just updated every single day, automatically, in a lot of detail, and then it has been updated to Obsidian. It's that simple.

OMI is a free app that runs in the background, listens, it watches your screen correctly, writes down what you do, say and care about all day. And it holds thousands of memories. Then you've got Obsidian, it's a three notes app, it's a filing cabinet on your machine. OMI exports into it, your agents log into it.

It's just a folder of text files that anyone could read it. And then what I like to do personally is I did it into our agent operator system. Cool. I did it.

Then it's edited to see your light. It's pointed at the same Obsidian vault. Every one of them wakes up already knowing me, with one memory across every agent. And so it's like, imagine this is something that, you know, when you think about it, it's absolutely mind blowing.

If you had a team of 50 people, an agency, think about how much time you'd have to spend training them up, coaching them, giving them, uh, knowing lessons, quality control on all their work, giving them a step-by-step operating procedure. And each person in your team knows something different about you. And each person in your team has their own weaknesses about what they know about you. They all have certain amounts of context, but it's almost impossible to know what their context is.

And how much they understand about you. Your business, your goals, and what it should be doing. And yes, you test them. But let's say, if you know what it's going to do, I do a lot of energy.

It is the best way for you to manage manually. Now, imagine that you could take your brain and you could just give your exact brain, all the stuff that you know, all the skills that you have, all the experiences that you've had, all the projects you're working on, all the contexts that you have across everything, and you could just plug that in to every single person in your team. And that is the best way in real life, right? Because it's not the best way to do it right now.

Maybe in the future, depending on how AI goes, that might change. But my point here is like, you can't do that with people, but you can do that with agents. And that's the powerful setup about this. And that's what changes everything when it comes to this whole system.

And so what I like to do is put it into this HLS. And then if you ever want to connect with the agents, like for example, going to Hermes here, and you can say, okay, make sure that you're connected to my Obsidian vault, something like that, right? And we just have to make sure it's connected locally to our Obsidian vault. And that's how easy it is to link any of your agents that are already located, even if you're Obsidian don't have the code, so that's how it works.

Now, it looks like I said, okay, this works for you, Julian, but this works for you, Julian, because you're working on all this stuff, or you're in the AI industry, or, you know, this is perfect for you, but it's not going to work for me, right? Some people say that, but every single person that I know has something that they're working on, that they care about, they have a vision, they have goals, they have the band, they have the voice, they have people in their life. And so if that's you, and it's just good, it's going to change everything. And I always used to think like, what's the point of having a second brain?

And then when I set it up, it really blew my mind in terms of how powerful it is. So when it comes to the Infinite Context Engine, you've got four layers and you have the loop, and this is the paradigm, right? You stop training your AI every morning. You start letting your AI train itself on you.

It runs in four layers, and there's a fifth that basically makes it infinite. So you've got the capture section. All you do is capture what you do every day. It's the third thing you do is taking and capturing your memories, and this will plug it into Obsidian.

Then you have the organization section, right? So everything that you're doing becomes clean, structured notes. Everything that Hermes does, you can log daily. How do you do that?

You just tell it to do that every single day, and you log that in as a scheduled task. So say you've called, all of these AI agents have scheduled tasks. All of these agents can know your schedule, update your HCDAs daily, and then you've got your goals, your projects, your clients, your habits, whatever you're working on, sorted into folders and linked together, so it's not just a big, messy pile of noise, it's something that's very, very organized and very, very useful, and then you've got store. So Obsidian holds the vault on your own machine.

Now, you can have this in the cloud too. So if you're using, for example, a VBS, and you've got an agent in the cloud somewhere, but I know there was a cloud just built recently. We've got training on that too. Wherever it is, you can have Obsidian linked to it, because even if you don't have your agents running locally, no problem, mate, you can actually have a login with Obsidian.

So you can actually log in with Obsidian, and then wherever your agents are stored, you can log in that to Obsidian, and then it's a shared memory of the multi-account that's everywhere. If you prefer, you can have it privately, no subscription, and it's a living knowledge base about you that grows every single day, but it's running fully local and fully private from everything else, and then deploy. So every agent reads the vault and starts every conversation knowing everything about you, right? So you can see, for example, you said, make sure you're connected to Obsidian, so yes, I'm connected to it.

I need to say, okay, just to test it right now, we'll say, knowing what you know about me, what are some amazing personalized SEO keywords that I could create content around? And I just want to show you the power of like, how you can basically use this for instant context, for instant personalized ideas. You can also use this in other ways. You could say to Hermes or Cord or whatever agent you're giving Obsidian access to, you can say this and that, make sure that you look at my work day-to-day and you tell me the number one thing that I could automate every single day, and then also when you do that, make sure that you just guide me through the steps and set up that automation, and so this way it becomes a recurring loop where Hermes every single day or Cord, and it's interchangeable what's got to be the agent, it's basically looking at what you did in a day, forming Obsidian a day, and giving you the best way to affect your day just to automate, so you can save time.

And then that loops up, it spirals up. So for example, Back in the day, I was doing SEO myself and now I have this fully automated. Now, the only reason I did that is because Hermes looked at my Obsidian Ball, it was like, mate, you just made it hours ahead of your date. I mean, he didn't say mate, but I think that's just an English thing, and I don't think Hermes is English.

But you get my point, like he can basically look at everything you're doing and be like, why are you spending all the time on that? If you just automated that and I could do it for you today, you would save so much time. And that's what AO Automation is all about. A lot of people waste their days messing around on old code and creating a vibe code and all sorts of crazy stuff that they never use.

Instead, what you're going to do is focus on the things that have the highest state pattern. And then also you can see, for example, here, even like for me personally, I spend a lot of time on SEO. We've got a bunch of personalized keywords based on what we work on day to day. So Hermes has looked at my Obsidian database, and it seemed that we work a lot on agent operating systems.

And that's something we spend a lot of time on. And so what it's done is it's actually broken down a bunch of keywords here that you can use for content. So agent operating system, AI agent operating system, Hermes agent tutorial, Hermes agent use cases. These are perfect, literally perfect keyword ideas.

And I would never get that if I was using another tool. So that's the other thing as well. Like you might have AI tools that you use. So for example, we use Ahrefs for SEO a lot.

But if I go into the keyword explorer inside Ahrefs, there's no option to say, you know about me. Give me some ideas. I could say, you know, check out my competitors and give you some key ideas, but it won't be personalized at this level. This is for you and you only.

You know, there's no other system that's going to be as personalized to you as this whole setup with Obsidian and Hermes and any other AI agent we use. And that's the power of this stuff. So if it keeps scrolling down, it's giving us some really, really good ideas on that. And that just demonstrates the power of Appendix BSS.

So there you have the loop as well. And this is why it's infinite, because every new chat logs back into the vault. Now that doesn't mean like every time you go back and forth with it, that just means you get a daily log from each of your agents and that goes into the vault. Anything useful should go into the vault.

So for me, for example, I have the Hermes article, I've spoken about it before. It's looking at the latest news and then it's creating my Obsidian vault on the latest news. Now, why do I do that? It's because when we're creating content, whether that's on social media, for SEO, for whatever, when we're creating content, it knows what's happened recently in the AI world.

And so it can give me new, trendy, relevant, personalized ideas based on the updates, right? You can see how it's doing this together. It's putting all sorts of things into that now that I've not even considered here. But the point is, it unlocks so many new potentials.

So you have capture with OMI, you have the organize, then you have store, and then you have deploy. That's how it works, right? Now you might also say, why don't I just use the memory system inside AES? Because AES has loads of different memory systems, like things like the Polychrome.

It isn't that I don't use those, it's because I don't own the system, right? For example, if you set up Poncho, you know, it's hosted somewhere else, it's something where I think eventually you have to pay per storage so that it can get quite expensive as well, whereas Obsidian is free. And also you can't customize it as much, whereas for example in Obsidian, you can customize it as much as you want, right? So you don't have to spend that.

So most memory setups are just one way. You feed the AI, it reads, and that stops improving the day you stop typing. The infinite context engine is a loop because every chat your agents have gets written back into Obsidian, automatically one file a day. So the Vault doesn't just store what you tell it, it stores what your agents actually work on.

Your ID, it's all pretty personalized to you. And then the next day, it reads that, it gets better, it looks better, and then it gets virtually legit. And your work trains your Vault, your Vault trains your agents, and this loops round and round and round. It's very important to understand this as well, because in 2026, like for me personally, I'm not prompting my agents that much.

I'm just giving them tasks, they go off in the background, and it loops around. You shouldn't be like having to babysit or manage your AI agents. Now how does that fit into Obsidian? How does that work?

The Obsidian, that is a loop system itself. It's looping round and round. Because the, for example, if you look at this, we've got the memories here that's logged in, and then we didn't do anything to save that, the loop did it. And so, it's you going to your agents, going to your Vault, going to your agents, and that goes back to you.

It loops round and round and round and round and round. So it's been something, for me, I organized to set this loop, but then I also use it when it comes to the chat with Hermes, as we saw a second ago. You also might say, well, this sounds like a lot of work to maintain, and it sounds like a lot of work to organize and create something as good as this. That's actually the whole point of this, it's fully automated.

Your agents are doing all the hard work for you, and you've basically got a full-time avid and early, outdated AI agents, who are the most experienced in different systems. Now also, what I love is the Memory Galaxy. So this is really the next level. If you don't have this set up, it's a beautiful system, number one, you know, I could sit there all day looking at this, it just looks fantastic.

But number two, it is a very, very interesting way to see how your mind works, because this is actually a representation of your mind. If this is a message, and I had like a misshaped galaxy, or for example, if I had stuff that was linked together that didn't even make any sense, well, that wouldn't be useful, but it would also tell me a lot about how I organize myself, and it would give me some self-awareness in terms of, okay, I need to be a lot more organized when it comes to managing this. So if your whole mind is a galaxy, you can open the Memory Galaxy inside your agent update system. Again, if you want to get my set up, it's inside the Airpuff World, link in the comment description, or go to theairpuffworld.com, but the main point here is like, you can open the Memory Galaxy, every node is a star, every link between nodes is a line of light.

Mine has thousands of stars, and it grows little by little every single day, right? And also, the stars that are shining the brightest are the ones that are the most recently updated. It's not just pretty, but it's actually very useful for having a look at this. Also, I think this is a much better representation than if I go into Obsidian.

And also, the other thing here is like, I don't want to be flicking between Claude, and then going back into the terminal with Hermes, and then wrestling with Obsidian inside a separate app. It's just way too much work. It's not going to be sustainable to get it finished, is what I'm thinking of saying. Because how do we do it?

It isn't for ourselves. We have to build a system that everything links together. That's what the game's about, my friends. That's how you get the most out of this.

And you know, even if your nodes, even if you've got like a setup right now, and it's messy, it's okay. Get your agents to organize it. You can go back into Claude and say, I've got some Obsidian files, but they need to be organized. The other thing that I'd recommend is to get the actual documentation for Obsidian, and you can train your agents with documentation from Obsidian.

So all the API documentation, everything else. This will just help your agents understand how a knowledge graph should work, how the whole system should be set up, and everything else. The other thing that I would say is there are some skills like Obsidian skill on GitHub. So if you've not created besties before, there's something called Obsidian skills.

It's got 400,000 stars of GitHub. And you can basically give that GitHub to Claude or whatever agent you prefer, and say install this so that you understand how to use Obsidian better. Now for me personally, I like to reduce the amount of skills that I install to my agents because number one, some of them are not that legit, so you always want to check the skills before you install it. And number two, the more tools and your experience you add in, the more predicated it gets.

But if you do find that your agents are struggling to use Obsidian, or it's not just working smoothly as you want it, this is a quick way to update it and get it back up to fresh. Now also something to note here as well, some people want to say, well, doesn't this use a lot of tokens? If your agents are recalling the memory every single time you work with them, doesn't that use a lot of tokens? So two things here.

Number one, I didn't want to be like, everything's about to fit, there's agents into the system. I would just have it do a daily sort of reconsolidation, right? So you can, for example, just say to Hermes, once per day, update the Obsidian ball. And that way you don't like add loads of crazy memories to it.

You don't like bog it down with too much details. You just add that one day summary, and that adds up every single day. The benefit of this as well, of course, is that those details never forget, get forgotten inside Obsidian, particularly when the whole conversation compacts, if it's trying to reduce the amount of tokens you're using. And then also, I tend to find that Hermes and all these other agents, when I go back and forth with them, if I literally put inside the chunk, it leads to Obsidian, it tastes a little bit harder, in terms of putting the Obsidian data seamless, right?

So we've got Omni, we've got Chats, we've got Notes, and Chats is in AI Agents. That goes into Obsidian, and that feeds Hermes, Claude, and every other CLI. So everything writes in, and everything reads out. It's a beautiful loop.

And so what this gives you back, if you look at this whole system, it's like, without a memory, it re-explains itself, the agents forget the details, it's the end of the road, get the tools that's required. You forget everything you did last week, and your agents definitely forget. And also, you actually waste a lot of tokens. trying to give more context and update your agents on what they're working on right whereas for example with this system it just knows so every agent wakes up knowing everything about you you won't be able to update any agent if the new agent comes out let's say for example open claw and hermes don't exist anymore but a new agent comes out let's say it's goldie agent or whatever we can just slot that obsidian bolt straight into it so you're kind of future-proofed as well because in the future there will be new updates you know open claw only came out in January and it's already there it kind of existed a bit but it changes ridiculously quickly more than any other industry that I know and so there will be something new that comes out in the future and when that new things comes out you're ready to just slot in the memory system and and that's just gonna help you save a lot of time and then you can find any past memory in seconds and it makes it smarter now also what I like is that you can search for it over here so you can search all our memories and be like you know like let's type in SEO we can see all of the memories that we've got around that particular topic so if I'm like okay well what progress did I make this year on SEO I can look through all the talks here if I'm like okay what have I been working on recently with SEO it's all a bit it doesn't have to be SEO I'll just give you an example because you have a SEO agency but you know SEO is good everywhere and this is my thing okay but isn't this expensive but OMI is free obsidian is free agents like Hermes for example are free as well and the only thing that's gonna cost you is a bit of time to set up and so that's the whole system now again if I was you I just want to show you you know what I do step by step if I was to set this up from scratch you know download obsidian and download and then you're gonna set up OMI to collect your memories and update obsidian vault daily and then you're just gonna go inside each agent and say hey connect to my obsidian vault locally if you do that you know that literally takes you 20 minutes but that gets you the 80 20 of where you need to be with this whole system and you can do that today but you don't even need 30 days to do this and so that's that's basically the whole system the animator the features inside obsidian so if you go inside the settings here as well you'll see for example you can update it regularly you can set up automatic updates if you're working in a different language you can change that inside your settings as well and then you can actually change like and customize all the different behaviors so the editing and that sort of thing you can change the appearance, you can add back links as well you can add a canvas you can change the default location for new canvases you can check out the daily notes section here also got a file recovery this is quite important to note as well it's like you know you can you might want to keep a backup of this every so often because this is storing like thousands of notes it's studying each bit so you have to come back to that later right and if you do for example only have this set up locally then if something happens to your computer then you can't access those notes anymore so definitely recommend just creating a backup like every 30 days or every quarter just so you've got all of your notes set up and you can come back to them anytime in case anything happens as well.

Now in terms of sync, so you'd add a login again I was talking about how you can add a login and then you can have obsidian synced everywhere then you can have a remote vault here and you can sync that so if you've got this on a separate device or you got multiple devices linked together then you can do that as well and that's basically it for the whole system that's how it all works you know you've learned a lot today you learned how to use OMI, obsidian, that's useful, how to use it, how to set up, how to get this whole process and also how to connect all your agents to this and why it's so powerful. So if you want to get the infinite context engine ready to go I mean you could wire this together you know but if you want the whole thing done inside our agent operating system as well so yeah a whole agent operating system and everything else with the OLE exports, the Galaxy F3 pre-connected you get that inside the AI, the full agent operating system, zip file, the memory loop, the Galaxy already wired in, we have four video tutorials on the OMI and obsidian setup, you get coaching calls four times a week where you can ask questions and set up the memory for other people, we have a room of 3,900 operators running this exact stack and the prompt CSP, it'll learn that for you so feel free to get out inside the FFT learning and look through the links in the description of the FFTlearning.co. Here's what I might say, okay this stuff sounds technical it sounds like hard work to set up we've actually got so many people I mean we've got a hundred and ninety-eight pages of wins and testimonials from people learning and growing and just having an awesome time and the thing that I would say is like if you said if you spend a day setting up obsidian like that could be quite a learning gap. You have a community where you learn, you win, you grow together, you share all your goals, you meet amazing people, you can connect with me personally as well, you can direct message me anytime and you can ask any questions as well so this is all inside the AI Profitable Boarding, we've just got pretty much an endless squad of amazing testimonials from people using the AI So you get the idea of AI Profitable Boarding, inside the community you can actually ask questions every single day and I personally answer them with a video tutorial so I run through all the letters questions and I answer them personally myself so if you need any help you can and then also people sharing how they've used their own systems here so for example like you see Shida may actually created a new team for obsidian with the AI Profitable Boarding, combining obsidian learning, combining the Lenton and Agents, that's the sort of stuff we share inside the AI Profitable Boarding and you won't find that anywhere else.

Inside the classroom you can get access to all of our best trainings, you can get access to all our best Lentons and you can see for example we have new daily tutorials every time something useful comes out so for example we actually just dropped a new one-hour course on how to set up agent operators for students, for agent OS so we update daily with a video tutorial you can see when it was last updated you get the video guide and this is it file to install it and then inside the calendar you can drop a weekly coaching calls and you can see this in your own time zone. Inside the map you can meet people in your local area who are building this AI Agents like Obsidian it's an amazing learning system, it's the agent that has to grow their business as well, yes we had a builder inside the AI Profitable Boarding, link in the comments description or go to the AIProfitableBoarding.com and again like if you have a look at this particular project this is pretty cool as well I mean again this is free and open-source it works with all your agents and it's basically just another lecture so these two do similar things. Something else that we have experimented with over itself and this will be pretty fun is you can actually work with wearables which is really interesting so you can actually sync this to your wearables you can see they do a bunch of wearables here and then you can have your wearables connected to your memory ball and that would connect to Obsidian as well which I think would be excellent. That's probably what we're looking at in the future I mean you can see the open-source AI wearables that some people wear in here, very futuristic idea and I've not tested out I do know some people that have done that and it looks super fun and yeah just super powerful, lots of fun so excited to see it in Graphify.

This is more for like AI coding assistance, Graphify is a really cool idea but it would just work on like for example understanding code bases so if you're thinking about using Graphify and you've probably seen a lot of people compare Obsidian and Graphify and everything else this is more for just getting knowledge graphs of one single GitHub so if you get a lot of projects like this one and you can actually visualize it with Graphify but it's just for one-to-one projects from what I've seen it's not specific for using full memory database and that sort of thing. So thanks so much for watching I appreciate as always I will see you on the next one cheers for watching all the way and let me know how you get on the ODE Obsidian and RH4, I'd be really interested to see what you can do with it, I've seen some amazing visualizations there man like you don't have to use the same UI that we use so I've seen some people like change the colors on this, change the style, change the way it's animated, there's so many cool things you can do with this and I'm excited to see what you can do with it so thanks very much for watching