ChatGPT Work 1 HOUR COURSE: Automate Anything!

ChatGPT Work Full Course One Hour: How to Build and Automate Anything.

So today, we're going to be running through exactly how to use the new app, which is ChatGPT Work. And we can download it like so if you want to get started. So you can get it from openingAI.com/chatgptwork. And we're going to be walking through exactly how to use this new tool step by step, some of the best use cases for it. It's a really powerful system now that GPT 5.6 with Soul is built in. So, you can do all sorts of amazing things with this app. Very impressive. One of my favorite apps right now for AI. And I'm going to talk you through exactly how to use it, how to get started with it.

We actually created this because we had a question from Jason 4 hours ago who was saying, you know, "Hey, I'm I'm using the new ChatGPT app. I like Soul. I don't know how to use it like the old Codex app." And to be 100% clear here, ChatGPT Work is the new version of the Codex app. So previously, this was called Codex, now it is called ChatGPT Work. And we can get access to it over here, as you can see. And this is what it's going to look like for you. So once you got started with it, once you've downloaded it, you just need to make sure you're signed in. And once you're signed in, then we can start using this.

So the first thing that you're going to see is you're going to see the new task at the top, right? So you can see over here, you've got "New Task" and over here you have "Chat". So there's two different versions inside one app. You've got the chat, which is like you can go back and forth inside a chat with ChatGPT. And then you can actually build stuff. Now, depending on what you want to do, you've got two different options right there. I'm going to walk through the Codex side of things first. So this is the new task section, as you can see. Um, from here, what we can do is we can just make sure that we have a new project set up. So what you'll see on the folder here, and you can click off it and then choose a new project if you want to. So we can click on "New Project", "Start from Scratch", or we can use an existing folder. Now, you would use an existing folder if you wanted to start a new project. However, if you don't have something new, uh, sorry, an existing project to change, then you can go to "Start from Scratch". So, we'll click on "Start from Scratch" and a super basic example. Let's start with the basics. Let's say, for example, you wanted to build out like a basic app. And you can build apps, tools, games, websites with this, whatever you want, really. It's a very powerful tool.

So, from here, what we're going to do is we're going to type inside here "habit tracker" like so. And you can see that we can switch between working locally or creating a parallel copy of our project as well, or we can work in a cloud environment. So "work locally" is you work on your Mac. So, for example, I have a Mac Studio. If I click "work locally", that means I am building on my Mac Studio. If we select "new work tree", that is going to actually clone our existing local project and we can start building inside there as a new version of our existing project that's been cloned that we can edit later. And then we have "cloud". So "cloud" means that we're working in an environment in the cloud that's sandboxed away and the agent can can work directly there. That's good if, for example, you wanted to share the task or you wanted to sandbox the task. And you also see this "usage remaining" section. So what this "usage remaining" section means is that you have a limited number of tokens that you can use inside your app. And the more you use and the higher efforts you dial your model to, which I'll come on to in a second, the quicker you're going to go through your usage. And your usage depends on which package you're on. So if you're not sure which package you're on, you can "learn more" here. You can find out what Codex plan you're on and you can go from there.

So what we're going to do now is we are going to tell it exactly what we want to build. And before we do that, we're just going to select the model that we want to use. Now, there's two options for selecting the model. The model is the brain behind ChatGPT. You can change and customize the brain behind it as much as you want. So, for example, if you wanted something really fast and lightweight that doesn't use many tokens, but probably wouldn't give you such great outputs, then you can actually change this to "terite". If you dial this up and you drag this forward, the higher it goes, the slower it'll be to respond, but the better responses you'll get and the more intelligent the effort levels you're using. And there's actually three models with uh, the new GPT 5.16. Soul is the frontier model, the most powerful. I would recommend if you're building something good, then you want to go with Soul. If you were just using the chat, basically, then you go with "terite". So we're going to dial this up to "extra high". I think for you watching this, "high" is probably good as well. And you can also switch to "advanced". And if you switch to "advanced" here, you can switch between the models. So Soul, Terra, Lunar, 5.5, 5.4. These are older versions of ChatGPT. And then we can change the effort level as well. And we can change how fast it replies. So, for example, if we want this to reply fast, if I'm doing a demo, which I am today, then I will switch this to 1.5x speed, which means I'll use more tokens, but I'm going to do that for you just to help you and so you can see results faster today. But for you using this normally, you know, this can code in the background. It can work on multiple projects and you can just go with like the basic settings here, as you can see.

So, we've got the "fast mode" switched on. We have "Soul high" set up. And then what we're going to do from here is we're just going to add, uh, you know, we're going to start building this project. So whatever you want to build, you start going from here. Now, you can actually click on these different presets as well. So you'll see, for example, "fix issues and failures". Good, if you have something you're already working on that you need to debug. Then you have "review code and suggest changes". So you can actually get the AI to look through what you're creating and then it can go from there. And we also have "build a new feature app or tool". And then we have "explore and understand code". So you can switch between each of these as you go along.

So from here, I'm going to go to "build a new feature or app". I'm going to say "build". And you see how we can build a feature, a UI, a prototype, an internal tool. We're going to build a prototype. And I'm going to say "habit tracker". That's fun, easy to use, gamified, etc. You can say little things like "beautiful UI". You actually, with the new models with ChatGPT, like back in the day, you'd be very detailed in the prompts that you give it. Nowadays, you don't need to do that so much. You can keep it simple and you can go from there. So we're going to hit enter like so. A little tip here, you can press Command and Enter, and that will just enter it for you so that you don't need to click this button over here. You can also control the AI with your voice. So if you click on "to dictate", then you can actually speak with your voice and we can ask it to do things like that. It will need access to your microphone, but we can allow that now. And you see how it's recording my voice. And then if we click on "stop", it will actually translate what was said. So you can see it's actually translating over here. So you can type with your voice as well inside ChatGPT.

And now you can see here that it's going to ask a bunch of questions, right? Says like, "Okay, before I build, I need three choices." And these are basically forks that help the AI understand exactly what you're building, how you want to build it, and everything else. So, for example, we've got "where should the prototype run?" And so, what I'm going to do for here is we can select what type of app we want to build. So, for example, if I'm building an app for like the iPhone store, then I'm going to go with like "native mobile app". If we go with "mobile first", this is like a web app, but it's going to be responsive first and then desktop is secondary. And then if we wanted to build an app for our desktop, then that's going to be the fastest to test, but it's not going to be native to mobile. Now, if you're wondering, okay, what does native mean? Well, for example, if we have a look at the agent operating system that we've built with all our agents plugged in, and you can build stuff like this if you want to using this system. Well, all these custom automations, they were built for desktop first and then it can be mobile responsive later. But the point here is that when you're building with this stuff, you want to help it understand exactly where you are building for.

And then we have the "style to use". So you can change the vibe, the feel of this, and everything else. So we could go with like "dark neon", "calm", "organic", "pixel game", uh, "playful editorial", etc. So I'm going to go with "playful editorial". And we can just copy that. Then we can go over here and we can say, "Okay, I want a responsive web app." And again, if you're building like a mobile app, then you could do that later by selecting "native mobile app". And also, all these things can be edited later. From here, we can switch between what type of prototype we want to build. So, do we want an "interactive demo"? Do we want a "functional local app"? Do we want a "full product foundation"? Now, the difference between this is like an "interactive demo" is like, you know, pretty simple, pretty fun to use, quick to build, but it's not going to be as functional as a "functional local app", right? Where you can, for example, like save stuff, you can create and edit things. And then we also have the "full product foundation". That means people can log in, they you can have a database, you can store details for any anyone who's using it, and they can save their progress as you go on. So, for example, you might start off with like an "interactive demo". If you like it, if you're like, "Well, this is good. We could build something out with that." Then you could build later the "full product foundation" for you. Because I'm doing a quick demo in a course, I'm just going to go with "interactive demo". So, I'm going to hit enter like so.

And now that it's thinking, and Soul is going to start coding. Now, one thing to note as we go along here is that you see how it's got the "full access" here. We can change the approval. Now, some people watching this will be like, "AI coding agent sounds great. Bit scary that it's coding locally for me. What if it does some crazy stuff?" So, you can actually change the approval levels here. So, you can switch between "ask for approval", "approve for me", or "full access". "Full access" means it's not even going to ask you for changes. That's usually the default that I go with 'cause I'm an advanced user. But for you, you might prefer to "ask for approval" just so you can check things and understand things before you allow it to do anything else. Uh, you could say "approve for me", right? So only ask for actions detected as potentially unsafe. That could be, for example, like deleting stuff, editing existing files. If you've got an existing project you've coded out, you might not want Codex to make any crazy changes. You might want to approve everything before anything goes live. So you can also switch with that.

And now you can see that it's planning the whole strategy. So what you can see is it's reading all these MD files. MD files are just simple markdown files. Very simple way of the AI understanding skills. So, for example, it can read this MD and you can click on it over here. And this allows you to understand what it is doing, right? What's going on here? So if we have a look, for example, at the "typography", this is basically like a skill that helps chat build things out. So, for example, we have the "typography" section over here and it has a pre-built skill where it understands, "Okay, these are the principles of typography. This is how it works step by step. This is what it works with. This is what it should avoid." Right? Mistakes to avoid. And it's basically like you level up the AI with skill sets. And if you have certain things that you want to train the AI to do, what's really good is you can create a skill. We can come on to that later. But you can see it's reading through all these different skills. It's understanding how to build them. It's also figuring out, "Okay, where should it build this as well?" And then it's creating the environment over here.

Now, the beautiful thing about coding with ChatGPT Work in 2026 is that when you're building with this, you don't need to be an expert. There is no technical barrier. You don't need to understand HTML. You don't need to be a coder. And don't let the Codex title sort of put you off. What this essentially means is that ChatGPT Work can code for you and build out technical projects for you without you needing to. You just explain what you want. And you can see it's now running through the separate steps. So it's got multiple steps inside the project that's creating. It says, "Okay, first we're going to do the responsive prototype. Then it's going to implement the habit completion. And after that, it's going to audit accessibility and responsive behavior." We can also add another step here. So we can click on "plus" and we can add another step as well.

So on the right hand side, you will see that we can review the whole project. We can see what's being done and we can see all the files being created by the AI in real time. First time I saw something like this, it was really off-putting, a bit confusing. I was seeing code and I was seeing all these files. Again, you don't need to worry about any of that because when you're building this, Codex will sort it all out for you. Pretty simple and easy. So, we have the separate files that we were looking at before. We can actually click off the review as well. So, if you click the X on the right hand side, if you don't want to see that anymore, you can click on it. And then also, if you want to expand anything, then you can click it like so. You can X off that. You can allow it to do web search. You can also see where it gets its sources from as well. So what it's doing to build this. And then we have the local file. So we can actually switch to Claude at any time. We can see the usage. So it switched from like 99% left over the next 5 hours to 96%. So it used 3% in the space of doing this first test for us.

Now, what it's doing is it is working on each of these, as you can see here. So it's finished working for 5 minutes. And what we can say is just, you know, "How is it going?" or "Can we preview this?" And we can click on "review" here and see what's working, right? So I'm going to say, "Can you open this?" Or so you see how it says "run it with here npm run dev". I want to say, "Can you run it for me?" And it'll begin to think and it will start the server. Now, if you're wondering what the development server is, basically, when it's creating something responsive like this, like a no local web app, it has to run it locally somewhere. And so what it's going to do is run it on a local port. So if we click on "open the habit tracker", now we can see the habit tracker on the right hand side. So if you ever see like, for example, it says "run it with" and it tells you to run terminal commands, don't run it. Don't worry about that. Just say, "Can you do that for me?" if you're not technical. And it's basically as simple as that.

So now we have the first preview of the app itself. And so we've built out a full responsive web app in the space of a couple of minutes. We can now open this up. So we can open up inside ChatGPT, which we've done over here. We could open it up in Google Chrome or Safari or Google Chrome for testing as well. So if we open up in Google Chrome here, you can now see that it is open directly inside my browser. So we can click off the tasks. We have this app ready to go. It's added the gamified sections that we talked about before. We can take things off and this is looking really cool, really interesting.

Now, you might look at this and you're like, "Right, okay, how can we add my vibe to the existing app?" So, let's say, for example, you're like, "Right, okay, well, I'm looking at this. So I'm like, 'This is pretty cool.' But what I actually want to do is I want to make sure that when I'm building with this, it follows my brand guidelines. It looks and feels exactly like my type of website." So, what we can do, for example, is I can go to the AR Profit Boardroom with existing brand colors like this. And I can open up the code from this website. So I press Option Command and U if you're on a uh Mac. If you're not, you can just right-click the page, click on "view page source", select the details of the website and branding that you want to emulate. And then from here, we can go back into ChatGPT. We can paste that in like so. So, paste it in as a long document. And the great thing about ChatGPT now is that when you're building with this, it's got a large enough context window that it can actually understand and take everything that you've got here, even though it's a really long document, and it can process that to use it for inspiration.

So, if I'm like, "Okay, can you make it look plus feel a bit like the attached HTML as this is my branding?" And now what's going to happen is it's going to analyze the HTML and then use that in terms of the typography, the borders, the spacing, the motion, the colors, the brand language essentially. It's going to use that to create a more relevant, personalized version of my style. Now, if you really like what it's doing here, if you're like, "Okay, I want it to do that every time and I don't want to paste it in every single time." Something you can actually do here is you can create your own skill file. So as we were talking about before, the markdown file that we were talking about previously, you can now use that for future projects. So we could save save this skill once it's done that and we can say, "Okay, can you save this branding style as a branding skill MD file so that when I create pages in the future, you follow my branding guidelines and I can call this skill file to change the style?" Right now, also, you can see here that when it was making those changes, the page went down and then went it went back up. That's because ChatGPT is editing the page in real time. It's making it better in real time. So, if we hit enter like so, we can go with that. We can also do, um, these different commands which I'll come on to in a second as well. But in the meantime, we will just add that as so.

So, that's beginning to build out now. As you can see, it says, "Yes, I'll package this as a reusable skill." So we asked it to save the branding style which is emulated over here as an existing skill so that in the future we can use this branding style and make more personalized, uh, projects in the future. So it's beginning to build that out now. And if we have a look at this newer version, we've got that ready to go. Now, this is still being used directly inside Chat Gepy, but what you'll see at the top here is the local URL. So we can go back over to Chrome or whichever browser you prefer. Um, we can have a look at this and now we can see our updated project with the branding style that has emulated my existing website. So instantly, we can import our branding guidelines and make it look and feel exactly how we want to do it. So you can see here that it's got the scrolling banner. It's actually customized the brand to the boardroom, which is part of our AI Profit Boardroom style. We have the colors that look and feel exactly the same, as you can see right here. And then this still works and functions exactly the same. It just feels like us. It feels more relevant to us. And this is really important because if you're building projects out in the future or you want to make sure they all look and feel like yours, fantastic.

Now, what we can also do is if we go back to ChatGPT here, we can see that our skill MD file was created. So this is the one that we have used for AI Profit Boardroom branding. So it's basically written a new skill file. Anytime that you have a workflow, let's say, for example, you do SEO, with SEO, you could save SEO as a skill and then recall it in the future as well. So the great thing about that is you can come back to it later, you can save it right now, you can reuse it in the future, and it's really easy to come back to whenever you want to. And that means that every time you do a workflow that can be repeatable, for example, SEO, might be creating videos, it might be writing blog posts, anything like that, save it for the future so you can come back to it later. If you want to change anything in here, then you can tell it, "Okay, avoid this in the future." And it can update the skill file for you as well.

Now, what we've got now is a skill file, a full project created. We've coded out. We didn't need to be able to code. We've got it hosted locally. We can test it out. We know how to open it inside our Chrome. We've made a lot of progress in the space of 21 minutes. What are we going to do next? So now what we can actually do is we could connect this to Netlify because if you wanted to share it with the world, well, how can you do that? I would recommend that you use Netlify. And Netlify is a way to host your website so you can come back to it later. If you have an existing website, you can actually give the ex access details to ChatGPT and it can host it for you. So let me show you an example of this. So Netlify is where we host all our websites. We can log in here. So we're just going to log in to Netlify, as you can see. And you can actually host projects on here for free if you just if you just want like a demo website. But if you wanted to host it on a customized domain, like so, then you can easily do that as well. And then what we're going to do over here is we are going to go to the "personal access token" section. And what we can do is we can host our projects on Netlify. Now, I'm going to have this expire, uh, very quickly. So we're just going to have this set for 7 days. We'll generate the access token as an example. We'll copy that. We'll go back over to ChatGPT now and we'll say, "Here's my Netlify access token. Please deploy this to Netlify for me, this website." And now it can begin to use that. Now, it may ask for extra documentation. And obviously, this is a quick example that I've created. So I will delete that in a second. But as an example of this, it's now deploying the validated website that we just created to Netlify. And then we can get a URL for it and share it with the world, which is fantastic because right now it's hosted locally and obviously you can't share local projects.

So these are some of the things you can do with it. Now, you can also have multiple different projects running at the same time. Let me show you an example of this. So one thing we did here is as a test, we actually got Hermes agent, like you can see, and we got the details of Hermes agent, which is a free open-source agent that we can use and run locally. And we gave that to ChatGPT so that it could start running Hermes agent for us. So it says, "Start Hermes with this." And I'm going to say, "Can you open plus run Hermes for me?" And if you want to run an open-source project with Hermes agent, for example, or whatever you see on GitHub and you're like, "I want to set that up but I don't know how," you can just say, "Install this, make sure it works," and you just hand it the link of whatever open-source project you want to create. So that's another example of how we can use this.

Now, if you're wondering, okay, where did that previous project go with the habit tracker? That is on the left hand side. So, you see how we've got a folder here that says "habit tracker", and then we have over here another project that says "Hermes agent". So, we can have multiple different projects being coded out side by side on the right hand side, as you can see right here. And we now have Hermes open inside a terminal window over here. And it's configured it with GPT 5.6 Soul. So we can now use AI agents. We can get Codex to install AI agents for us. And whatever you want to build or run locally, we can go with. Now, you also see here that in the meantime, with our personal access token from Netlify that we asked ChatGPT to deploy for us, we have that available here. So if we open up that URL, we now have a sharable URL. This one right here that we can share with the world. So, if you create something cool and you're like, "Do you know what? I want to share that with other people in the world," here's how you can do it.

Now, if you go back to your projects inside Netlify, and you wanted to add a custom domain to that, you can actually click on "domain management", "add a domain", "buy a new domain", and you can find a new custom domain for your example, right? So, we can type in "best habit tracker". We can search for domains. And then if you wanted to host this instantly on a custom domain, you can just pick whichever one you want. So you've now learned how to deploy the project as well. We can also delete the personal access token. So, for example, if I go back to my personal access tokens here, I'm going to delete the project that we just built. As I've just shared that on a video, we don't want the whole world having access to my Netlify account. Then we can delete it and we can come back to it later. So pretty cool.

And what we could also do from here is we could use something like Superbase so that people can actually log into our website. So, for example, if we have a look at this, you see how there's it's kind of like a functional demo. It works, but you can't like log in. There's no login page. There's no option to create an account. So if we wanted to link this to an app where we had a database where, you know, you could share this with customers, they can log in, you could use it for lead generation, etc. Then you can go over to "plugins" here and you can type in, uh, for example, "Superbase". Now, Superbase is a system where we can install a database to whatever projects we create. So if we log in with a free account on GitHub, we can authorize that now. Allow ChatGPT to edit and write for us. That's fully created. Now we can go back to the habit tracker over here. And also, one thing to note is you can pin the task or you can pin the project. What that means essentially is if you have a project that you're like, "Coming back to every single day," and you don't want it to get lost in the list of projects that you have right here, you can click on "pin project". And with the "pin project", you can now basically come back to that later and it won't get lost. It won't get lost in the feed. It won't just drop down to the bottom. It will stay at the top. And we can drag that to the top as well.

So what we can do now is we can type in "@" to tag in the app that we want to deploy to. So if we have a look at this, we're like, "Right, okay, I want to deploy this to Superbase." So I'm going to say, "Using Superbase, create a database where people can log in." Now, it's not going to have access to my Netlify anymore because I've removed that. So it will start running locally again. I'm going to say, "Make sure you just do this locally for me." And we, you see how when we give a command, it's waiting. So what we have to do if we want to update the existing chat here, it's a really useful feature. Then you can click on "steer". And "steer" means that whilst you are speaking to your agent, you can steer it in the right direction. So you can say, "Okay, do this, go over there," etc. Right? So you see how I said, "Make sure you do this locally for me." Steer. So we interrupt what it's currently doing and we steer it in the right direction in case we forgot to instruct it. If you see it's going a bit off course inside the chat, which can happen with AI, but it happens less and less. Now, then what you can do is you can steer it at any time. So now it says, "Understood, local only. I won't create, restore, or modify any cloud Superbase project. I'll add a local Superbase database to it. Wire the habit tracker in." And it's now going to build that inside Docker. Okay, so it's now beginning to use that directly.

Now I can say, "Okay, deploy it to my local project, but use, uh, Superbase. You can use the cloud version of Superbase." Then we click "steer" and we guide it in the right direction as well. So that's one example. If you wanted to, for example, look at this app and you're like, "Okay, I really want to make this into a system that is an iPhone app, for example." Then we can go back into ChatGPT and we can say, "Turn this into an iPhone app instead." Hit enter and steer it in the arrow direction as well. So whilst it was designing the off-gate with Superbase on the habit tracker project, we steered it in the right direction. So we're going to redesign the whole project to be an iPhone app instead.

So those are the basic ways that you can build out projects with Codex. Now, what you can do over here is you'll see there's a bunch of other options. So, for example, we've got "scheduled", we have "plugins", we have "sites", "pull requests", and "chat". Let's start with "chat". There's a big difference between the project section here inside ChatGPT Work and the new chat over here. And so, for example, if we're looking at this, we've got our recent chats. Now, these are actually pulled in from ChatGPT. So if we see "all", we can see all of our previous chats with ChatGPT in the cloud. This actually syncs with our main ChatGPT account. So if we go to chat.openai.com, we can have all of our conversations on the left hand side. And if we go back to ChatGPT Work, we now have the chats from the web app directly inside our local version of Codex. Fantastic. So now we can go back inside these conversations. We can actually add those conversations to a task as well. You can look through your history. You can create a new chat. And so if you just wanted to use ChatGPT instant, you could say, "Okay, what time is it?" Now, just as a totally random example, but just to show you how you've got this different version.

Now, you might be wondering, okay, well, when do you use the chat and when do you use the ChatGPT Work features inside Codex? So let's just call this "Codex part" for now and this part "chat". So with the chat feature, this is just good for going back and forth and getting quick answers in between projects. So it's kind of like you can use ChatGPT and you can use Codex at the same time. So you can build out projects, but you can use the power of Codex as well at the same time here. Now, also, if we go back to the conversation and we can minimize the chat. Now, if we go back to this conversation, it said before converting it, I need three decisions. Now, you might be sometimes you might be in a rush or you don't really know the right answers. Now, at this point, a GT 5.6 Soul is very powerful. It's a very powerful frontier AI, but it's very easy to use and it's designed for non-technical people. So, what you can actually do here is you can say, "Okay, figure it out and handle this for me." So if we type in that, it can actually decide for us and make decisions on our behalf. So if we are not sure of the right choices and you need to consult with GT 5.6, you can ask it, you can get feedback, you can, um, problem-solve together. The same, for example, if you see things that you don't like inside the app, you can ask it, you know, "Change this, add this," um, you can customize anything as much as you want, which is really, really powerful.

So we also have the Hermes agent over here too. So when we are working, we can work on multiple different projects. We can multitask at the same time. Great for getting multiple things done. Not great for focus. So if you need to like be really, really focused and get something very complex done, then I would recommend that you just stay inside the chat here and you focus on one project at a time. But if you're trying to be super productive and you know you're building some big stuff out and then you're switching between all these different things and you are you don't want to s stand there waiting for like 2 or 3 minutes for it to finish. As you can see, it works for like 1 to 2 minutes each time, or sometimes 5 minutes at a time. Well, then you can have multiple different projects working in parallel and you can switch between them. So if we want to build on the habit tracker, we have that here. If we want to switch over to Hermes agent, we've got that here. And we can switch between this much as we want. So with Hermes agent, for example, we could go inside and we say, "Okay, can you set up image generation with GPT 5.6 inside Hermes?" And we'll see if it can do that. And it might say no, or it might say yes.

Now, also, that is one important thing to note here is that when you are building out with ChatGPT, it has some advantages to Claude and other AIs that you can do. So, for example, if you look at ChatGPT, you can generate images, you can, for example, use web search. And Claude itself cannot generate images. And if you want to generate images for Claude, often you have to use the API, which can get expensive. With this system, instead, you can use GPT 5.6 and the power of it on your existing subscription without paying for an API. And so, for example, if we want to generate images inside the chat, totally random example. We could click on the chat here. Let's just test this out. So, we are going to say "create image". So we click on the plus sign here, "create image". And then we're going to say, "Okay, create an image of a dragon in a castle." And we can generate images on the side here whilst we're working on these two projects. So we got Hermes agent, we have the habit tracker, and we have the image generation inside the chat over here. Pretty cool.

Now, when we're using this as well, so whilst these two projects are building out, so we got the live tests with Hermes agent for image generation. We have the setup with the habit tracker as well. We also have some options for more usability features. So I'm just going to minimize that. And also, we can add more apps to ChatGPT. So let's say, for example, you were working on spreadsheets. You could actually allow ChatGPT to edit and create spreadsheets for you. So you can click on "try now" and you can just allow it to create spreadsheets. You could also go back to the plugins here and you can install whatever you want. Now, the more tools you add, the slower the response is, and also the more time-consuming it can become to reply, and also the more tokens it uses as well. So you want to get that balance. Now, for example, we can have a look at all these apps and be like, "Okay, well, let's say, for example, I want to manage my emails." I can get ChatGPT Codex to use it right here. So I can click on "install plugin", "continue to Gmail". We can go over to Gmail here. We can connect our emails, as you can see. Click "continue". Then we can go back to Chat Gepy for Work. And it's now connected to my emails. Right? So we can say, "Okay, summarize my emails, draft replies, process email threads." And so one thing that I used to have back in the day was a virtual assistant. And I would, you know, hire them hourly. We'd get them to manage our emails, but sometimes they take weekends off. Understandable. Sometimes they get tired. Sometimes they don't follow your instructions. The powerful thing with something like Gmail inside ChatGPT, and you can do this for any workflow that you work on day-to-day, is you can actually get this to automate tasks for you. So I can replace a virtual assistant that would normally handle my emails with Gmail. And so, we can automate and save time, be more efficient. And the other thing about this as well is that this is a daily task. So managing my emails, my Gmail account is a daily task. Now, I don't want to go into Codex and ChatGPT directly every single day and ask it to do this manually. So what we can do over here now that we've added a plugin is we can go over to the "scheduled" section and we can create a new scheduled task. So that means that as long as we have ChatGPT Codex running in the background here, we can come back to Chat Gepy Work and it will run scheduled tasks in the background whilst we're doing other stuff.

So if we click on "new task" over here, um, we're going to say, "Let's set up a schedule task together." We're going to select a new folder for this. So we can actually have this running in the cloud as well. We can start from scratch on a new project. And we can change this name to "Gmail Automation", "Daily Gmail Automation". So, we'll hit save there. And then we're going to say, "Let's set up a schedule task together." First, explain how scheduled tasks work. You can do that if you want, but I'm going to delete that bit. And then I'm going to say, "Let's set up a schedule task together. Manage my @ Gmail account on a daily basis at 4:00 AM every day so that all my emails drafted before I wake up." Right? You can see the instant obvious benefit of that. We save time. We have an automation. We don't need the team. We don't need to manage everyone. And we've got it running on a schedule that I don't even need to worry about anymore. Now, you might be wondering, okay, well, why would what happens if the agent sends something crazy or or goes wild with this stuff, right? So that's why we've said it, you know, "do the drafts first." And then what we can actually do is we can pin that project as well so that we come back to it later. And it's now going to use that tool that we just added in one single click, which is Gmail. And it can manage my Gmail for me every single day on automation. And you can do this with any tool inside the plugins. So if you, for example, need to create a presentation every single week to let's say your audience about a certain topic, you could say, "Okay, schedule a daily or weekly task where you do some deep research for 1 hour. You find all the latest news in my industry. You create a presentation about it so that I can present a video." You could say, for example, "Go through my Notion and start posting a blog every single day about certain topic." You you could say, for example, "Every day, review my habit tracker and based on the habit tracker we've built, make it better every single day." So you can see how you can create these loops that run in an autonomous cycle without you.

So we've also got the image generated here as well from Hermes agent. And we can ask Hermes at any time, "Generate an image of," and it will generate images inside our AI agent. Now, one of the interesting things about this is that, you know, if you went back a year ago, I'd have to install Hermes agent myself. I'd have to be the one who's problem-solving it. I'm the one who has to run through the setup and the installation and everything else. Now, with GPT 5.16 Soul, it's so smart that it can do all of that stuff for us. And we now have image generation as a tool set up inside of Hermes agent. We have the habit tracker with the setup right here. And actually, we've got the habit tracker set up as a mobile device here. I'm going to go back into ChatGPT and I say, "Allow me to preview the app." And it's going to start using it here. So, we've been pretty productive so far. We've set up an agent with Hermes that can generate images for us as well now. And we've got an AI agent ready to go. We also have the habit tracker prototype. We've switched that to mobile. We've added the database and people can log in and they can check that out whenever we want. We can actually open that up on iPhone whenever we want as well. And we're good to go. We know how to use the chat over here and we know how to automate our Gmails as well.

So, you can see now that we have the schedule task for Gmail drafts running at 4:00 AM every single day. And here's what it's going to do. It's going to review my emails. It's going to read relevant thread context. It's going to avoid duplicate drafts. It's going to exclude like the newsletters and everything else. It's going to create the drafts inside Gmail. It's never going to send, delete, archive, or label emails, but we could ask it to do that if we're more comfortable with it. And we can report anything that is drafted and flag anything. So now what I can say is, you know, "I want to test that schedule task. I want to make sure it actually works in the background." So I'm going to say, "Okay, test it now." And so what it's going to do now is start looking through my inbox with Gmail and run the schedule task. Now, if you want to come back to your schedule tasks and see what you've automated, then you can see that over here. So, if we go to the schedule task section, we've got a list of all of them, and

You can build these up. You know, if you set up one automation every single day running on a scheduled task, by the end of the year, you got 360 automations set up and you've pretty much automated every single thing inside your personal and business life. That's how powerful this can be. And it's a system that runs and improves every single day. It is something that gets better and better over time.

Now, we have the habit tracker app ready to go. So, we can preview that in Chrome. Here's the app. As you can see, people can now log in before they start using this. We can create an account. We can type in our name. So, let's test this out. Add a password. Click create account. And just check this out. Let's have a look. So we can enter it now and our habit tracker is built. Pretty cool.

And then we can come back to Hermes agent. Let's just test that out as well inside terminal here. So I'm going to say create an image of a cat skateboarding. And you can see this working in the background.

Now, we've also created two drafts inside our emails for these two responses here. And this will just run on a daily task. Now, it skips some of the messages that we don't want to do, don't want to mess around with. And it creates these. And so, we've automated our emails in the space of 2 minutes.

Now, that will not be perfect. It will not write in the tonality that you want every single day. It will not do things exactly how you want it. So how do you improve that? Well, you can say, okay, make, let's say, for example, the responses were too formal in our emails. And you would check your email inbox to check this and make sure you're happy with it. Or you can say, okay, send me the email draft that you've done so far. Hit enter like so. That's going to start thinking and it's going to start reading the emails sent and pull that in so that we can have a look at it.

We can also ask it to send emails for us as well. So, we can say, "Okay, send an email to juliani@gmail.com saying hello." It's like, "Sure, I'll send an email to julian@gmail.com right here." So, if we pull up our emails now, we have the email as you can see. So, we've now automated our emails. We've taught chat how to send them and we've got them handling our emails every single day on automation. And if you wanted it to actually like send emails live view, you change this. You would say instead of running on the schedule drafting emails, send them instead.

Let's check Hermes agent as well. Now, so it's created the image using the GP5.6 automation that we talked about earlier. And what we can do over here is we can go back into Hermes agent and just say open up and we can get it to open our image for us. Now, it's also edited Superbase and everything else.

Now, some other cool stuff that you can do with this. If we go back to plugins here, we have computer use. This is really impressive. So what it can do is it can actually control your computer, but it does it in the background without interfering with what you're currently doing. So if we click on try now, we can say create a post on school.com and I'll send it the link to our school group. We'll go back into chat about the best ways to use chat work. And so we've told it exactly what URL to post at, what to post about, and to use computer use. And now you can see it's actually working in the background. To this little popup over here is computer is working directly. So if we go back to our Chrome here, you can see it's opened up a new tab. Now, this is my mouse moving side to side. This is the CodeX computer use agent working over here.

Now, if we go back to the chat, it's actually given us a plan title and what it's going to do. I'm going to say, okay, talk about how to use it for SEO and include a link to a free SEO strategy session. Something like that. Or you could say, I've created a 1-hour course inside the app for boardroom on how to use chat GPT for work. Add that as a CTA inside the post and we'll add a link. Use this link and we'll include some other information about what it is. So, you know, you could add your testimonials and that sort of thing. So, I'm going to get it to post a value-driven post about how to use chat GP to work and then how to get access. So, it says, I'll add the CTA at the end. Perfect. Let's do it. And now it's going to add the post. So, it asks you for your approval. It doesn't do anything crazy without you saying hello or you saying okay, do this, don't do this, blah blah blah. And now we're going to get the computers to post on school for us. And this works really smoothly or anything. You can see how smoothly it works here. So it's created the full setup here. And now it's posting for us with this gray cursor. If we open up the post, pretty nicely written value-driven post. So I did a nice little prompt on how to use chat work. It's kind of worked like magic. And you can imagine like if you were writing that, it would probably take about 5 to 10 minutes. If you're posting it, it's going to take a lot of clicking around which uses up your energy and that would also be quite time consuming. And then remembering to add the CTA as well is better. And also, you know, if you had the choice between like a virtual assistant or, you know, hiring someone on your team for doing social media for you or for example just getting chat cheap and chaty work to do it for you, which one would you do? I would go with this one.

And also the cool thing about this is we can say, okay, schedule that in every single day. So do this as a scheduled task every single day on something new and relevant in the AI industry and you actually schedule that task in and then we've got two automations built in the same morning. So we have the Gmail scheduled automation and we have the school post automation if we wanted to build that. So we say schedule a new post in and do it every morning. And now we can create that schedule task. Great thing about that is as well that people forget to do stuff. But if you tell the AI to do it, it can't forget. As long as you keep it open, you've got your desktop open ready to go, then it's going to have this as a scheduled task. It'll just run like clockwork. And if you want to test it, you can do that anytime.

Now, computer is really cool. Building out the apps, really cool. Setting up new agents and new GitHub projects, very fun. Automating your emails, absolutely great. Saves a lot of time. Now what you can also do is you can start using this to, for example, like create videos for you as well. So you can see here we said based on this skill, create a video about the AI profit boarding for our website. So we've taught it the skill of using Remotion. Remotion is another GitHub skill where it can basically create videos. So this AI agent can now create videos for us. So if we go back to chat GPT now, we've got the full details here and we can say, okay, create a new video every single day. This is like everything that you want to do on a daily basis, you can do with this process.

If you want to watch the actual video that we've built here, we go. So that is the image and then we have the video over here. So, this is a video fully animated and fully created using chatchec CodeX as well. And so, if you're like, well, how do I automate my social media? Well, you can get it to post for you. You can get it to create videos for you. You can get it to post on a schedule for you. You can get it to automate your emails. Really simple and easy. I mean, for example, if you were doing lead generation or lead outreach, you could actually schedule chat GPT work to reach out and generate leads for you on a daily basis as well. And the video itself is nicer and better certainly than I can do. Better production quality as well. And the other thing you could do here is you could actually link it to something like Hey Gen, for example, and then it can generate videos with your AI avatar as well, which is absolutely mind-blowing when you think about it.

This is an image of a cat on a skateboard that Hermes generated that before as well. Look at the quality of it, the colors and everything else. Looks super nice and that was fully generated with Hermes agent that was set up by GP5.6. And the great thing as well about using GPT 5.6 on CodeX is that instead of using APIs, you can use OAuth login. So instead of paying for GPT 5.6 soul API or Fable 5 API from Claude, you can plug that OAuth login into Hermes agent, into OpenCLaw, into any other agent that you use if you're into that sort of stuff.

So productive morning and here's an example of a website we actually built for the AI profit board as well using similar systems. So this created a website for us in one click using this whole system.

Now, you might be at this point wondering, okay, well, I've learned so much. We've learned how to basically link this to any app so that it can do stuff in the real world. We've learned how to schedule those tasks, so it's kind of like an AI employee that just runs 24/7 for us. We've learned how to automate anything and how to build apps, games, tools, websites, whatever we want using the system. You might be saying, "Okay, what do I do next?" So for you, what I recommend is that when you're using all this stuff, look at where do you spend your time on because I've seen so many entrepreneurs in particular who love this sort of stuff and it is fascinating. It's super exciting. I have loads of fun building with this sort of thing. But the problem is a lot of people, they get stuck and they're like, "Okay, well, I'm building out all this stuff and I'm spending all this time on AI, but I don't know how to be really, really productive with it."

So, what I recommend for you personally is that you write down all the tasks you're working on day-to-day. You track your time, what you spend on it, the tasks you're doing, and also if it's repeatable or not. And if you're wondering like, okay, well, I don't want to write that myself or how do I track that? So, you can just use like a basic timer or you could actually create a new project inside CodeX for the time tracker. So we could say create a time tracker where I can track and audit everything that I work on day-to-day and what I need to automate. So it's like a simple app where I log the task, how many hours I worked on it, wherever I'm doing it every single day or the regularity of it and how much time it took plus the benefit of it. And this way you can easily build an app where you track all your time and then based on what you spend your time on, it's like, right, okay, I spent three hours on video creation today, that's what I need to automate next. Or for example, I spent 3 hours posting on social media today, okay, I need to automate that next. Step by step. And then how do you automate it? You create a new project and you tell CodeX exactly what you're working on and how you're doing it. You test, iterate it, and then that just cycles, that just loops it around and around.

Now, another superpower of using CodeX as well is something called forward slash goal. So goal mode, as you can see here, you can set up a goal here and you can say, okay, continue this task to completion. So you see how it says goal at the bottom, you just type in for/goal, hit enter, and then you can say, okay, build out a beautiful website and don't stop until it's ranking on Google. Build out a beautiful video and don't stop until it's absolutely amazing. And what that's going to do is cycle round. So for example, you can say, build out a website for the AI profit boardroom and don't stop until you've got 50 blog posts created for SEO that are related to AI automation with Hermes agent. So, we've got that goal. Now, what you can see here is that it says, I'll turn this into a tracked goal and build it through to a verified site. I'm using the brainstorming app. So, you know how I just told you to track your time, track your tasks, track what you're working on, and then start automating each part of that. You can use forward/goal mode to run that. And you can see right here. So it's like, okay, what's the CTA? So CTA equals this. And you can be really short with your answers, like if you don't have much time. Hermes agent equals an AI agent that you can learn about in the AI profit volume. The articles should be about 1,000 words each. So, we're now going to tell it, okay, on the website, link to this page. Here's what Hermes agent is about. Here's what the articles should be about as well. And it's going to start pursuing that goal. And this, I've seen some people literally run this for days. So, you can have an autonomous AI agent that runs without you and just focuses on that goal.

This is great as well because for example, if we go back to the habit tracker tool here, it was kind of a little bit too back and forth. We don't want to be prompting it like that all the time. We don't want to be going back and forth with it and having to check it every 5 minutes and then telling it what to do and making all these decisions if you're creating a lot of stuff and building something big. So what you can do with this instead is you give it the details of the project you're working on and you say, do not stop until this is done. And how this works is it loops around. So you have the goal that's working on an autonomous loop and then you have a judge that actually looks at the goal and figures out, okay, is this goal completed to the quality or not. So it's a self-improving loop and it will just loop around as much as you want until the job is finally done. And so that will tackle the task. It will take a few hours, but the job will get done and this is so much more powerful than if you were, for example, like just going back inside the chat here.

So these are all the things that you can do with it. You know, we've covered new tasks, scheduled plugins. The sites feature here is basically like you can build out your own websites using this stuff. And then we have, for example, pull requests and the chat plus how to build anything inside CodeX. How to see your previous projects over here. The tasks as well. So we have all these tasks that are being worked on in the background. So for example, when we gave it the task of doing chief of doing the school post, you can see that the task is over here, right? So you got the different tools that you can use, the pinned projects, the projects that you've worked on previously that are not pinned, and then you have the tasks that have been worked on as well. And you can minimize any of these. Well, if you click on them, if you want to clean it up, you can do it like so. And then there's not loads of stuff on the sidelines here. And you can also drag this and resize it as well.

And then inside your settings, if you click on the bottom left here, there's something called a pet. Now, pet shows you everything that you are currently working on. So all the recent chats, all the things that you've recently done. Uh, so if we click on the pet, if it's working on a task in the background, for example, like this one, you can go over to Chrome, you could be doing something else, and the pet will tell you what's going on. And also, if you click on it, it will take you straight back to chat work, as you can see here. So, it's a really powerful app. Super useful. GPT 5.6 soul, especially, is unbelievable. Really powerful. If you ever want to stop one of these tasks, you can click on stop over here. And that's basically it. That is how to automate and build anything with CodeX.

And again, this was thanks to Jason who actually just, you know, asked this question. If you want to ask me questions like this, you can check out the AI profit boardroom inside the community. I create a video tutorial like this answering everyone's questions every single day. Inside the classroom, you can get access to all of our best trainings for this stuff. We actually have a full 1-hour course on how to use GT5.6 as well. And we have a full agent operating system that allows you to use all this stuff inside one amazing dashboard. And then we add new daily trainings based on what's actually useful. And you get a video tutorial and a step-by-step guide here as well. Inside the calendar, you can jump on weekly coaching calls. You can ask questions, share your screen, jump on a Zoom meeting, meet other cool people doing this sort of stuff. And inside the map, you can actually connect with people locally near you who are building amazing stuff like this too, using agents like Hermes agent, using CodeX, using agent operating systems, etc. And inside the profitab um, you can direct message me as well. So feel free to get that link in the comments description or go to the apiprofitboarding.com. Ice watching.