The AI Chief of Staff System I Use Every Day

If you're using AI in your business right now, I am going to bet that you're jumping between a bunch of different tools. Chat GBT for this, quote for that, maybe a note somewhere with some prompts, another tab with your project context. And every single session, you're starting from scratch, reexplaining who you are, what you're working on, and what you decided last Tuesday. And no matter how many conversations you have, it never actually gets better at helping you.

So, I built a system that does. An all-in-one AI workspace that has access to all my knowledge, my work, my tools, my context, and it just gets better the more we work together. I'm E, and I help knowledge professionals turn their scattered note systems into a personal AI knowledge system where everything works together so you can move faster with more flow and focus. And in this video, I'm going to walk you through my entire setup. We're going to look at why I switched to using Claude Code and the app I used to run the whole thing so I don't have to touch the terminal. The one folder that holds my whole system and controls how it behaves, makes decisions, and shows up for me. The system that gives it real memory across conversations, so every session picks up exactly where it left off. And I'll go through the tools and skills that I'm using to level up the system and turn it into an actual operational partner in my business. By the end of this, you are going to see exactly how to go from a scattered AI setup to one system where your AI assistant works alongside you as a real operating partner.

Now, before we jump in, if you're a knowledge professional who wants to turn your scattered notes and tools into a personal AI system that can actually operate beside you in your business, then check out my membership, College of Knowledge. It's the place for solo operators who want to punch above their weight with systems that operate at the level their mind already moves at. First, you'll get access to the full vault of resources. All my courses, templates, and workshops that walk you through building an all-in-one system step by step. Then, join our live workshops and office hours where you can get direct help from me and the community. And talking about the community, we have an amazing community of solo operators just like you who are obsessed with this stuff. So, if you're done with scattered tools and disconnected workflows, check it out at the link in the description below.

All right, so before I dive into the actual system and the files and folders and all that kind of thing, I want to just talk to you about what an AI system actually is, because it is fundamentally different from just chatting with Claude or Chat GBT. Okay, so every time you start a chat with one of those systems, you are essentially starting from scratch. There's no context. There's nothing. You might have some custom instructions that you put in, but essentially the only thing it knows is what you are telling it in that chat. When you start to build an AI system like I'm going to show you, you start to plug in different components so that when you start a conversation, it already has some context built in. It can pull in knowledge. It can pull in data. It can pull in skills. And so rather than just like having to explain yourself every single time, it's actually kind of a system that pulls what it needs when it needs it. Okay?

So I think of it in four different parts. The first part is context. And when I show you how to use kind of co-worker code for this, this is kind of like one file. We call it the Claude.md file. And it basically explains who you are, how you work operationally, how you know how your business runs, how Claude can actually show up for you. I'll show you my full Claude file and like some of the things that I put in there that actually make Claude a partner for me and a chief of staff for me. So that's kind of context, right? And that gets loaded in every conversation so that you're never starting from scratch again. Okay.

The second part of this is skills. And I had like now over 50 skills that I have built so that my system knows how to work for me. And these are literally just repeatable workflows. Okay? So, I built workflows for a morning brief that my chief of staff does for me, content operations, weekly planning, all of that kind of thing. It's all codified into skills, and those skills just help your AI to know how to do things for you. So you don't have to keep repeating yourself week after week and that kind of thing. Okay.

Then the third part of an AI system is the tools that you plug into it. And when we have a look at kind of using, you know, upgrading your system to some of our codework or code, this is where you can start to really connect tools, right? So we connect these through MCPs. You can also use APIs if you're using Claude code. And so it allows you to go from just chatting to actually having AI operate for you. And this was like a massive light bulb moment for me because I went from like chatting with Claude, you know, being a power user for AI for a long time, but when Claude actually started to be able to do things in my tools, I was like, "Oh, this is it." So, um, tools are really important.

And then the fourth thing and the fourth layer of this AI system is your knowledge and your data. So plugging it into, if you have a knowledge system. I personally have a knowledge system in Tana. So Claude has access to all of that. It can go and find things. It can go and discover what I've been thinking or all the content in a certain topic. And I've also built some dashboards with Claude that accesses different data. So it accesses all my sales data, my membership data, marketing. So I can actually ask Claude different questions about things that are going on in the business and it can go and find the information. So that is kind of like how to think about that's a massive upgrade from just chatting with Claude and doing a few little things to actually having a system that works for you and your business.

All right, so let's talk about the actual system. Okay, so I do not use Claude chat, as you can see here. Okay, so I have totally switched over to Claude code. And the reason for that is as I kind of explained, you know, when we're talking about the AI system, is that chat really just is back and forth chat. You can't really build that all-in-one system just with chat. So what I'm going to show you can be set up in co-work or code. And what I want to say is Claude code is not as scary as you think it is. It's also doesn't have to be just for people who are, you know, developing apps or things like that, right? The reason that Claude code is actually so powerful is that it can actually connect into APIs. It can build things for you. And you might not think like, well, I don't want to build apps. That's not what it's going to build for you. I built business dashboards for myself. I've connected it into API tools so it can actually do work for me. Like I run a soap community for College of Knowledge, and I actually have a full skill that can actually go in and it can upgrade people, downgrade people, it can do all sorts of things. And code can do that because it's connected into APIs and MCPs. Co-work is still a very good kind of like middle ground, and everything I show you today can actually be done with co-work.

Now, how co-work and code work is that when you first come into one of these, you will choose, in co-work it's called a project, but in code it's called a directory. And so your whole AI system lives within this one folder. And so everything I'm going to show you today, I'm building within this one folder. I call it Chad. Chad's my AI chief of staff's Dave. And so you start there, and basically it can access everything within that folder. It's just a folder on your computer. Right now, I don't actually use the Claude app either. I have another app that I use called VS Code. I like VS Code a little bit better. And there's a couple of different reasons for that. Right. The first one is that I can actually have multiple tabs open. And I often do this. Is I'm often, you know, chatting in different kind of tabs, and it helps me to get work done faster. So within VS Code, you can actually open up different tabs and have different conversations running, and it runs on the exactly the same thing. It runs through whatever plan that you have, Pro, Max, that kind of thing. It works within this app. So VS Code is absolutely free. You can download it, and once you do, you just want to come into extensions and install a Claude Code plugin. Okay? And that will just get you up and running. It'll look exactly like this.

The second reason that I like VS Code is because it has this file editor. So you can see here, up here it says Chad. That is actually the folder that lives on my computer. So if I come into the finder and go to Dropbox, Chad is in here, right? So everything that you see kind of here in the file browser in VS Code is just a folder on my computer, right? And that's the whole system. It really just is a folder. And so I can come in here and I'll explain kind of some of the things that I do a bit later in the video, but this is kind of our projects folder. So I can come in here and I can have a look at, you know, documents. I usually switch those to preview. And I can view and update documents. Claude can update those documents. And so it's a nice kind of like all-in-one operating system for using Claude for me.

So let me show you through the folder that has everything that is in this AI system. So every time I open up Claude, it has access to this whole folder. It can access anything in it. So the first thing that you're going to want to do is you need a Claude.md file. This file gets loaded at the beginning of every single conversation. So this is your context. This is whatever you want your, I call my system Chad. He's my chief of staff. And so I'm going to call him Chad from now on in the video, right? So Chad needs to know things about me. It needs to know who it is. It needs to know how to show up. All those kinds of things. So every time I want to have a conversation, all the things that I want it to know is in this Claude file. So if I open up Claude, and let's open it in preview. Okay.

So there's a few different things here within the file that make it work really well for me. Now, it's up to you how much or how little you have in a Claude file. Just remember that it does add to your context. So, just I try to keep mine kind of lean, although it's kind of growing and I need to fix it up again just to get it a bit tighter. But essentially, I have at the top who Chad is, right? You are Chad. You're my chief of staff. You work alongside me. And then I kind of have a very small paragraph about Chad's personality. And I just write this as like how I need him to show up for me. Okay.

The next part of the Claude file is just about me. There's a whole lot of stuff in here. By the way, I did not write a single word of this whole file. So, if you're looking at this and thinking, uh, I don't have time for this, don't worry. Right? I am going to leave a link to a prompt. I call it the personal contact profile prompt below. And basically, Claude will interview you and write your Claude file for you. Okay? So, you can see here just a little bit about me. There's all of this stuff here, routing rules. All of this has come from the work that we've done together. Okay? So, as Chad has worked with me and, you know, let's say we've had a conversation and something hasn't worked out, he updates his Claude file, right? So, like, and this is what I mean by it gets better every time you work with it, because Chad just now knows if something goes wrong, he's got to remember that and he puts it here, right? So, there's a whole lot of different stuff in here. There's stuff about, you know, I use Tana. There's stuff about Janet, who's my, she's my AI editor. There's things in there. There's some stuff about, you know, tools that I use, that kind of thing, right?

Then we have the next part is the main skills that we have. So it just kind of lists some of the main ones. I find this good to to be included just because these are the ones that we use all the time, right? And then down here it says other skills are in the skills directory. So when you mention things like that, it can go and look in the skills directory. We also have basically an automation and tools registry. And essentially when I say registry, it's just a file, right? It's just a file that says here's all the tools that we have. I'll show you what that looks like when we get to tools. But essentially in this Claude file, what I'm doing is I'm pointing it to other files that it can go and look at if it needs. Okay? And what I found with a Claude file is that it's much better to mention other files and folders and directories than have it all kind of in one. Essentially, what we're doing is saying, if you need that, go and look at it.

All right. Now, there's a bit more about what I do here. This one here's really changed how Claude works with me. I put in a whole lot of stuff about my personality, my human design, my disc profile, everything. And so now the responses that I get are really, really different to the normal kind of responses that Claude would do. And I found this to be really helpful. So one thing about me, and if you're familiar with human design, I'm a generator. And one of the things or ways that I work is that if you present me with like an open-ended question, I find it very hard to answer. And I much prefer when I'm presented with options, right? Cuz I can respond to an option rather than just kind of like open-ended. I'm like, ah, blank page. I don't know what I'm doing, which, you know, just a personality quirk, right? So, I've actually baked that into the Claude file so that every single time we're working either on content, on strategy, on any kind of thing, Chad will actually present me with options and I can say yes or no. I've just found it so much better. So, having kind of personality quirks, how you work, and essentially again, this is how your AI can show up for you in conversations. So, that's been really helpful. There's work style here. So when I work best, communication, planning, my word of the year. So it kind of knows kind of quite a lot. This current context because I work with Tana, Chad actually just mapped my whole Tana workspace. So it has kind of like shortcuts to different Tana nodes. So it can get there pretty quickly. And then I actually have a whole lot of details about its vault. So it actually knows how it works, how the vault works. And by vault I mean kind of the system. So it knows the structure. It knows kind of how memory works. And I'm going to talk about my memory system, which is one of the best things I've built. It's got rules. And like again, I'm just showing you this, it didn't start this way. Chad has literally just built this as we go. And as I said, it probably is getting a little bit too long. So one of the things that I try and do kind of with the system every month or so is actually just go back and we optimize it. So it's like, okay, the Claude file is getting way too long. We need to work out how we can shorten that up. We just want efficient, especially with like context and memory. You just want it to be efficient because now this is getting down to kind of like about almost 500 lines, and I want this to be at about 200, right? So there's a lot of things in here that don't need to be in here. And so what you want to think about with your Claude file is essentially like, it's loaded into every conversation. So what is important for your AI assistant to know in every single conversation, and what can be moved out so that it can grab the information if it needs to into the conversation, but it's not actually in the context in every conversation. Okay.

So, the best place to start if you're getting up and running with a system like this is to download VS Code, get the Claude Code plugin, and literally just come in, start a conversation, and say, "Hey, we need to start a directory, build a folder on my computer, and start our Claude file, right?" And then Claude will actually just go and do it, right? That's the one thing I love about Claude code is that it just goes, it does things. It updates files. It can, you know, move files. It's beautiful. Right?

So, back to the Claude file, right? The rest of how it works. Let me come up here. It's a bit harder to see in the directory, but you'll see what it looks like. But essentially, the structure of this file, right, of the structure of how the system is actually made up is there's a folder called daily. That's for our daily logs. So, essentially at the end of every conversation, Chad logs what we did. I have found this to be helpful for myself, right? I've kept daily logs and a daily page for over 5 years. And I said to myself, could this be helpful for my chief of staff to also keep daily logs? And turns out it has been. So that's part of our memory system. We keep a folder called threads. This is kind of for more like bigger projects and things that we're working on. And so Chad kind of keeps a thread on each of the things that we work on. And I'll talk more about that later in the video. We keep a folder called assets. So, if we're working on the things that actually, you know, we're creating together, that's in the assets folder. There's a whole lot of scripts, okay? So, Claude can actually create things on your computer, right? Little, you know, scripts and things that do things. So, I've got scripts that schedule my content for the week. I've got scripts that send me a daily brief every morning, and Chad just goes and creates them. And that's why I think that if you're a solo operator, upgrading to code code will actually upgrade. Like you can make things happen, and you can actually just say to Claude, "Hey, could you send me a report every morning on where the business is at?" And Claude will just work out how to do it for you. And that's the kind of thing you need as a solo operator because you're operating on your own, right? And so scripts are kind of where we keep that. We also keep references, and I'll talk a little bit about what references are, but essentially they get pulled in the Claude file. If it needs to go and reference something, they're in our reference folder. And then also there's some dependencies, and that's just code stuff, and I just let Claude handle that.

So let me talk about the memory system that I set up because I don't use the built-in memory system from Claude, right? Because I found that it was saving little memories all the time, but they weren't actually helping to be able to consistently have conversations about my business. So what I did was I thought to myself, what actually helps me to, you know, be consistent with the things that I'm thinking about in my business? And there's two things. One is I keep a daily page. So in Tana, I have a daily log with everything that I'm doing, what I did for the day, all of that kind of thing. And I just keep that updated throughout the day, right? And that's really helped me to kind of like be able to, you know, go back through the week and see kind of like what I actually did, what was I thinking, you know, those kinds of things. And the second one is like, you know, just the notes that I keyed about different projects that I'm working on, where I'm up to, that kind of thing. And so when I was building out this system, I thought, okay, well, if that knowledge system works for me, maybe it will work for Chad. And it has actually worked so well. So we work on this two-folder system where we have our daily logs and we have threads. So in the daily logs, you can see here, we've been doing this for weeks now. And at the end of every conversation, I have a skill called log this. And so basically I just say, "Yep, okay, log this conversation," and it will log exactly what it did, right? So it's got all bullet points here, anything that we worked on. And I found this really, really helpful because how I kind of built into Claude system and in the Claude file, like if I mention, "Hey, we were working on this the other day," or, "You know that kind of thing," it will actually just go back through the last, you know, few days of its daily logs and it can then remember, "Oh, okay, catch itself up on what we were working on," and I can actually continue the conversation. It's changed kind of how I'm able to just continue on with conversations and have it remember what we actually did, right?

So, let me open up my skill and show you. This is the skill that I created for logging. Now, what I have done is I had Chad create a script that runs in the background. So essentially I say, "Log this," and it created just a little piece of code that reads through the whole transcript and then logs it into the data log. And that was not hard to build at all. All I said was, "Hey, I want to have you, you know, build the script that logs each session," right? And it just works it out. It builds the script, and now anytime I say, "Log this session," that's what it does. And it's worked so, so well. And so those daily logs, they just keep compounding over time. And like I said, anytime I say, "Hey, we were working on this the other day," it will just go and look through the log and catch itself up. Okay, so that's kind of like one part of our memory system. It's kind of, I would say it's more shorthand.

Okay, the second part of our memory system is what I call threads, and they are kind of key for me. So these are like bigger things that we work on that span multiple conversations. They span multiple days, they span multiple weeks, months, that kind of thing. And so whenever we work on something large, let's say it's a new client project or something like that, I'll say, "Hey, can we create a thread for this?" And essentially, I call them threads because they're kind of threads of conversation. So I have threads for a couple of different things. I have threads for like main things that happen in my business. So for instance, I've got a thread for my College of Knowledge community, and it has basically all of kind of the working stuff of what's in the community, you know, what we're doing, you know, those kinds of things. And it works anytime I say something about College of Knowledge, it goes and grabs that thread, catches itself up, right? So it works really well like that. I've also got threads for like big projects that I'm working on. So at the moment, I'm working on a new app. It's called HIO. And so we've got all the details of the app, what we're working on, what's been shipped already, what still needs to happen, those kinds of things. So anytime I talk about HIO, it's like, "Oh yeah, okay, it can just go and catch itself up with these files." I have files for all of my weeks, right? So whenever we do weekly planning, it creates a thread file. So throughout the week, as I'm updating it on what I did and that kind of thing, every day we do an end-of-day check-in, it can update and get up to speed on what's happening in the week and kind of what's changed, what's shifting, all of that kind of thing. And so all of these files mean that I can literally just continue conversation. So at the end of the day, I'll say, "Hey Chad, just a bit of an end-of-day check-in." And it will load the weekly plan. It will get itself caught up, and then we can actually just have a conversation with context. And I do this with large client projects as well. So I build like custom workspaces for people, and so I will start a conversation with Chad, talk about the strategy. We'll kind of like put some stuff together, and we'll create a thread file for it, right? And then, you know, I might have to send something off to a client or something like that. I can come straight back into a new conversation and say, "Hey, client next approved that proposal. Let's move on to phase one." Right? And then we can have a conversation about phase one. And so essentially, the habits that I've set up for myself is, and Chad is actually so good at this now, and it's because it's in his Claude file, is he knows the structure of memory and he, he knows the structure of the vault itself. So he actually now suggests to me, "Should we create a thread file about this?" So essentially, if we're working on something, he will say, "Is this a big project? Should we create a thread file?" Or I say, "Let's create a thread." And then as we keep working, let's say I open up a new conversation, as we keep working on it, I'll then say, "Hey, can you update the thread file with what we've discussed today?" And so it keeps us both on the same page together, and it has been the best memory system that I have built. I have all of the normal Claude memory turned off now, and this is the only memory that we use. And every conversation just feels like he knows exactly where we are in the business, what we're doing, what my priorities are. You know, if I mention a project, he knows what I'm talking about. It is fantastic.

All right. So once we kind of have memory and context and all of that built in, one of the other things that I have worked on a lot over the last kind of one to two months is building the skills, right? So skills are essentially repeatable playbooks that Claude can do every single time. Same kind of process, that kind of thing. And so I essentially listed all of the things in my business. I don't know if you see this up here. This is like, like I just listed all the kind of things that I'm doing over and over again in the business. And each week I kind of like do it with Claude and build the skill for it. Right? A lot of people are just going into the internet and trying to, you know, find and download a whole lot of skills. I've done none of that. I actually just build skills from the work that I do every day. If I come in here and have a look at all of these skills, I think I've got almost 50, right? And they are broken up into a few different categories. So Chad has some internal skills that he has. So, there's the save thread. There's system upgrade, which he does for me every week. So, he goes through all of my bookmarks. If you saw last week's video, I'll put a link up above, but essentially I bookmark different things for him to look at, and he upgrades his system. How he can use Webflow, which is updating my website. We then have a bunch of skills for my College of Knowledge community, and we built different things like coaching follow-ups, codifying the work that we do here into workshops that I can share in the community, new member welcome workflows, office hours, workshop design, all sorts of things. We do a payment check every single week. So I used to log into like three different platforms to see like, are there payments that have been missed? Have people cancelled? Those kinds of things. And now I can do it in about 5 minutes. And Chad just goes and checks all of that. He goes into Stripe. He goes into Thrive Cart. He goes into Circle, and he actually just gives me a report every week on, is there any missed payments? He drafts the emails. It's fantastic. Right? So all of that can be built into things like skills. We then have a bunch of content skills, cuz that is, you know, one of the main things I do. And then down here we have operations and some Tana stuff. And I guess like how I think about skills is in a couple of different ways, right? So I think about skills in terms of the workflows that I do. So that payment check is one that I use all the time. The second type of skill is something like the Webflow skill that's up here. So this is a skill that is specific to a tool, right? And one of the things that I found is when I plugged in tools and MCPs, is that Claude still doesn't really know how to use them. This was very apparent to me when I plugged in Tana and also something like Webflow, is like it's just trying a whole lot of different things because again, it's a blank slate. It doesn't know how your system is set up, that kind of thing. So, a lot of the skills that I build are not just workflows for me and Claude to work together, but actually like how to go and do something. So, in my Webflow skill, it's got details about the site, how to access it, publishing rules, it's got the different kind of like tools that work better than others. It's got details about all the styles that I have. And what I find is that having a skill like that, and we unleashed it on my actual website. And then what I did at the end of that conversation is say, "Codify this into a skill so you don't make the same mistakes again." And that's essentially what I'm doing with almost every tool now, is that I'm building some kind of skill so that it's like a shortcut because if you ever seen Claude kind of work in tools, it's trying all sorts of different things. It's, you know, going over here and doing this, and I just was like, I just can't look at it use all of my usage up with, you know, trying stuff. So every time we kind of use a tool, I create a skill for that tool. I have a massive skill around that I call Tana Copilot because Tana is my main knowledge base. It's where it can get a lot of information. And so it's got information in my Tana Copilot skill down here. It's got like a full system map, like maps to all the nodes, everything that where it can find things, that kind of thing. And so rather than every single time I say, "Hey, can you go and see my quarterly plan?" And it's then would have to look up the quarterly plan and do this, now it doesn't. It's like a shortcut straight there. And so I love building those kind of skills for actual tools. And then I also love building skills for the work that we do together, right? So a lot of my content work is just kind of process. So for instance, let me go through this one. Most of my skills look pretty much like this. So what the actual thing is and how to trigger that skill, and then down here what the structure of that essay is, how to write different hooks for it, principles and kind of like what my worldview is when I'm thinking about this. There's some hook patterns, and then there's the workflow down here. So like whenever I say, "Okay, we're going to write an atomic essay," it will pull all the context that it needs from this skill, and then the first thing that we do is activate the idea. So we're looking at what mechanism am I teaching? How do I frame this for the audience? You know, those kinds of things, and we work through that together. It then presents me with a brief. It then goes and writes a draft. I react to that draft. And so, the whole process is actually built into this skill. And I find a lot of the things that I used to procrastinate on, I don't procrastinate on anymore because I got these skills, and Claude just walked me through the whole process. I'm still in it. I'm still, you know, giving all of my ideas, and I'm still driving, but it's actually driving the process for me. And I don't write any of these skills, by the way. So the way that I build all of these skills is we actually do it together in a conversation. Look, the first time we ran this skill, I would have said something like, "Hey, we're going to write an atomic essay. This is what an atomic essay is. Here's some of the examples from old essays that I've done. Here's kind of the general process I want to run." And so we will run it in a conversation first. And then at the end of that conversation, I will then say, "Can you go and create a skill so we can do this, you know, repeatedly?" And Claude is actually great at going and writing its own skills, right? And then the next time I run it, I do the same thing. It'll actually pull the skill, and we'll run it through. And then, you know, if there's things I didn't like or that didn't work or we need to update this, at the end, I'll just say, "Hey, can you update the skill with everything that we learned in this session?" You know, I did this with my weekly planning and probably over about three to four weeks, we actually locked it in. And so now I almost have to do no updates to the skill because I just use that iterative kind of way to like really lock that skill in. And so now I have kind of like, I said, over 50 of these skills, and for almost everything that I do in the business.

Okay. So the last thing I want to show you is the tools that I actually have connected. So one of the things that upgraded my whole system was connecting into the tools that I use every day. Right? So I started with connecting MCPs. Right? So MCPs are pretty easy to get connected. So even if you're in the Claude app, you can come in here to customize and you can add these connectors. They're called MCPs. Different apps have created a specific way for your LLMs to integrate with their tools, right? And that's kind of a great way to get started, right? And I have a heap of MCPs that are already connected. And then I kind of went one further, and this is why I really love Claude code, is because I then discovered, well, Claude can actually access different APIs. So even if a tool doesn't have an official MCP, I can give it the API key and it can actually go and do things. And this is like a massive unlock for me. Now, one of the things, if you are going to kind of get into APIs and things like that, is never give Claude an API key directly in the chat. Okay? That's very insecure, and you should never do that. You should treat your API keys like passwords. So, what I do instead is I keep all of my API keys in the Apple keychain, and it keeps them secure. Claude can still access them, but it accesses them in a secure way. So, I've got a whole heap of ones that are here in the keychain that I've added in. And essentially, think about it like this. Anything that can connect into Zapier, it basically would have some kind of API that you can grab. So all the different tools that I've got connected into my system is all of these, right? So I've got Tana, which is where I keep all of my knowledge, right? Tasks, notes, ideas, docs, all of that kind of thing. It's connected via an MCP. And essentially all this is, is this is just a file that I ask Claude to create that has kind of a map of, you know, everything that's connected to our system. I have Stripe connected, so I run my whole business on Stripe. And so I can ask Claude to go and, you know, grab an invoice, you know, an invoice link, or, you know, create an invoice, or do this, or check payments, or things like that. And it does. I use an app called Posters, and that schedules all of my social media, which is fantastic. It's connected via an API, and I love it. I also have a Claude built a tool for me so that I can schedule my Substack posts. We have Kit connected, which is for my email marketing. Telegram, Circle, which is where I run my community. I do some kind of work with Obsidian, so there's some stuff in there. I've got Google Workspace connected through the Google Workspace CLI. 11 Labs, which is voice and text to speech. There's X, Webflow, ThriveCart, Fathom, YouTube, everything. Right. So all of the tools that I use every day, I have connected in some way into Claude code, which means that I can do things like, "Hey, can you grab that transcript from that call I did the other day?" I can get it to go and look at my YouTube analytics. I can get it to actually update my website. These are all the things that if you're a solo operator, you're doing on your own. And so having that extra help with Claude is just chef's kiss.

And so having this all-in-one folder, right, where you've got your Claude file that tells you how it's going to show up for you today, the memory system with daily and threads, all the tools connected, all that kind of thing makes it so that I can say something like, "Chad, what have I got on for the week?" And so Chad can pull up what it needs to pull up, right? It's pulling up operational context. So that's from a thread file that we have. It's going into TANA. It's checking, you know, the calendar nodes. It's going to look at the week 14 thread that we have. And so essentially at the start of a conversation, it's pulling in the context that it needs so I don't have to explain. And then I've got, "Here's what we've got on," right? It's a half week before, you know, I go on holidays. Today, this is what you're doing. You're filming YouTube videos. Tomorrow you're doing this. Friday's a holiday. You know, this is where the cycles at. And all of that comes in because I've got this all-in-one system. Right? So up here you can see it's kind of pulled in the context that it needs. It's not necessarily that all of that context is all loaded. It's pulling in the context that it needs when it needs it, right? And that's from the thread files that we have. Plus also because it's connected into Tana, which is my knowledge system where I keep, you know, all of my tasks and projects and, you know, ideas and things like that. Now, if I asked another question like, you know, it's the end of the month. "Hey, how do we do in the business this month, you know, sales-wise?" I can actually do that because I've got it connected into my business dashboard. Next week, I'm going to be sharing a video on exactly that. So, make sure you subscribe to the channel to see that video. I'm not going to do that right now cuz I want to, you know, show every my sales stats. But, I can do all of those things now because I have this all-in-one system. And it really has upgraded how I use Claude in my business. I'm no longer just kind of like, you know, little chats here and there. I am using it every single day for almost every single conversation and every single operational thing that I do in the business.

Okay, so that's how you go from a scattered mess of AI tools to one system that actually works alongside you in your business. One of the workflows I run on this system every single week is a full strategic planning session. So I sit down, I say, "Let's plan the week." And 10 minutes later, my entire week is mapped out. So I made a video showing exactly how that works that you can check out here. And remember, if you want to build a system like this, that's exactly what we do inside College of Knowledge. So, I'll leave a link in the description below. And of course, if you like this video, hit like and subscribe so you can see more of it in your feed.