GPT 6 Astra AI Full COURSE 1 HOUR (Build & Automate Anything)

Welcome to this one hour master class on GPT-6 and Astro. I'm just going to show exactly how to automate basically anything using GPT-6 Astro, which has blown me away since it's been released. It's super powerful, fantastic to use, really fast, very, very impressive. Honestly, I never thought I'd see the day, or I didn't expect to see the day this year where I would switch from Claude to GPT, but that day has come, right? GPT-6 Astro is very, very impressive. So today, what we're going to do is we're going to run through a full step-by-step course on how to build and basically automate anything using GPT-6 Astro in this course. And step-by-step, we'll run through exactly what it is, why it's so powerful, how to use it, how to access to it, and how to basically automate anything with it. If you want to skip through anything, I've got the list of what we're going to cover today inside this document. So, you know, if you want to skip through for certain things that you prefer to check out first, you can skip through the timestamps, but we're going to get straight into this, all right? So, let's have a look here, and we'll have a look at some of the things that you can create. So, I mean, you can see here, for example, like, just from their examples, like, you can fill in spreadsheets, it can, for example, do game development. We can use 3D, MCPs and modeling, like, for example, Blender, which I'll show you, some examples of in a minute. It can do your basics, like formatting, documents, and quality control and everything else. You can also build apps, tools, games, websites, etc. In fact, I've built out full-blown, like, dashboards, like, you can see right here. This was, I literally cannot code, right? I cannot write a single piece of HTML, but I can create a beautiful tool, like, you can see right here, and the DIY looks amazing. And the way that I've done this is by using GB6 Astro to help me improve the UI, to help me add the functionality in everything else around it. So, it's an incredibly powerful tool, and with that, let's get straight into exactly how you can use it, all right? So, the first place you can go to is, you can go over to chat GBT. Now, this depends on what plan you are on, all right? So, you need to make sure you're on the right plan. If you're under in which plan do you need to actually use GB6 Astro and get access to it right now. And so, you see the details here, GB6 Astro is rolling out to limited set of organizations, and will them be available to chat GBT+, pro, business, and enterprise users, as well as through the API, right? You can also get access on Microsoft Azure and ADES bedrock. So, there's a few different ways you can use this, as you can see right here. And, you know, feel free to use it however you want to. For me, the two places that I'm going to be using it are inside chat GBT directly here, and also inside the chat GBT work app here, all right? So, those are two of my favorite places to use chat GBT, and I'll run you through the basics and how you can use each of these in a second, all right? So, now, you might be thinking, okay, where do you start with all of this? What can you do? What can't you do? What are the limitations, et cetera? So, if we have a look, for example, and we go to chat GBT and GB6 Astro, as you can see here, I have found literally no limitations. Now, we'll say that some things are better than others when it comes to these designs and that sort of thing. I'll cover what we can do and what we can't do. But, I mean, for example, if we have a look at this, this was a video that we generated with, I, fully edited, or fully generated, as you can see in this example, looks beautiful, it's better than most editors could create, and this was fully done with a free tool, a free open source tool, and GB6, and the other thing that blows my mind with all of this is that you can do it with one single prompt. So, if you want to get access, two ways, right? So, make sure you're on the right plan, which plan is going to get it on, plus pro business, and enterprise, where can you get it, you can go to chat GBT.com. And then, the thing that I've noticed here inside chat GBT is like, it's got a six pro, as you can see, if you scale up the thinking effort over here, or you can use higher, extra high. I think the default here on the models, if we check, is we've, but you can also go to work here, if you go to work, you can actually select, "astra" inside the scroller here, and my favorite place to use is actually inside the chat GBT app. You know, I think you can do a lot more with it, I think it's a lot more fun, and you know, if you're watching this live, by the way, feel free to ask any questions, as you go along, but yeah, I think on the sea, inside the codex app, it's much better. So, we've covered how to get access, let's talk about some example automations that I've built recently, and what we can do with this. So, we've covered video generation, as an example here, what else can we do with this? So, if we go over to this section, we actually created a full game with this system, right? So, you can see this game right here, right? This game, we created this game in pretty much one single prompt, using the combination of Blender and GBT-6 Astra. So, the reason that the graphics looks so nice on this is that you can have a generate images and turn those into games, or you can use this image generation alongside Blender, which is like a 3D model in software, and then you can basically get Astra to control that model for you. But the quality of that game is beyond anything that I've seen, anything else create, really, especially in one single prompt, you know. You might say, okay, should I be using, for example, Claude, or should I be using Gemini on that sort of thing? I mean, it depends what you prefer day-to-day, and where your setup is. But even for me, as someone who's been using Claude for months now, and really favoured it over Chat Cheapity, I will say that Chat Cheapity's Astra is smoother, faster, better at computing use, does everything that Claude can and can do it better, is also much better at being direct. And what I mean by that, for example, is like, if you ask Chat Cheapity Astra to start doing something, you just get straight to it, and it's very direct. If you ask it a question, you'll answer your question. And you might think, well, that sounds basic, like every AI can answer a question, surely. But actually, if you go over to something like Claude, Claude sometimes refuses to do stuff, sometimes it doesn't answer the question properly. It does seem to be very bloated and slow compared to anything I previously used with any other model, you know, with literally if I compare Opus 5 or Fable 5 or anything like that, it just gets distracted, right? Where I know that if I ask Astra to go and do something, it will just get it done as well. So the thing that I like when it comes to agents, particularly versus Claude, like Claude, you can't plug into Hermes agent and Hermes agent is probably like the best open source AI agent out there. Now the problem with that is like, if I can't use Claude with my subscription already paying for inside Hermes agent, I can't plug it in, but I can with GPT6, then I'm far less limited with GPT6, right? I can use GPT6 Astra frontier model inside Hermes agent, just using my assistant subscription, I don't need to pay for an API and file that. That is amazing compared to like not being able to use Claude and if I want to use Claude as a brain of Hermes agent or any other agent that I use, well, I have to pay for the API and that's kind of annoying, honestly. And it also pretty expensive, so I wouldn't recommend that, you know, like and these are some of the biggest reasons that I'm switching over to Astra. It's just, it's faster, it's more direct, it's smoother, it creates better quality outputs and also we can plug it into our favorite AI agents. Having said that, some limitations that you need to be careful of, I've mentioned already that you don't get it on every single plan and also another big limitation is the weekly usage limit, right? So I haven't used it like a massive amount, I wouldn't consider myself like an absolute power user of, it's the experiment in my life, only got access about two days ago, right? Two days ago. And that feels like a long time ago, actually. But when I look at my usage here, you can see we've already used 44%, 44% of our weekly usage and I'm only like two days in. So just be careful of that, be mindful of that. It's like, yes, fantastic, but it can use up a lot of tokens. So there are limitations to this. Having said that, super powerful, easy to use in everything else. So just to recap here, we've talked about examples of automations with video generation, games, I've shown you the agent OS tool that we've built with Astra as well. So you can build all sorts of cool, fun, wonderful stuff. If you actually go on X, X is a fantastic place to really see what people have created. And honestly, it's just so many amazing things I've seen people create with this tool, like it's unbelievable, right? And you know, I think that we're really, I mean, it's a little bit high peak, you know, like I don't want to hype anything up too much. But if she look at what, I think it was that video that came out and said yesterday, maybe Jensen, who I said, jeep to six is AGI. Now for me personally, I mean, what even is AGI, right? Like everyone has around sort of definition of it. At the same time, what I say is the most powerful AI I've ever seen? Yes. Is it the most impressive model leap? I've ever seen between GPT 5.66, absolutely. So yeah, I do think it's a fantastic model. All right, so we've talked about video generation, games, tools, websites, et cetera. Let's talk about how to automate anything and just get straight into the action. So let's say, for example, you wanted to build with GPT 6 Astra. You may be asking yourself at this point, okay, what should I build? I would start with where you spend your time the most. And so for me, for example, if I'm spending a lot of time on like generating ads, then guess what, I'm gonna automate first with GPT 6 Astra and add studio. Why? Because that's the thing that I'm spending all my time on. So, you know, whatever you use with Astra, don't try and automate things where you are automating for the sake of it, right? Automate something that you actually spend time on day today. And I know that sounds obvious, but you'll be surprised by the amount of like people who are running the business and just don't make stuff that they don't even need. Like wasting their time with things that aren't even useful. Don't do that, right? Just focus on the things you actually spend your time on and that will change week to week as you add more automation soon. So as you build more automations, you're gonna find that you're spending your time on new things and therefore you need to create new automations for those new things you're working on. So it loops up all the time. Loops up all the time. We've actually got inside the AI profit boardroom. If you haven't checked out already, link in the comments description or go to the AIprofitbomb.com. If you go to the classroom here, we've just added a new section inside the beginner to expert course that basically shows you how to identify your time, how to stay focused and how to never feel like you're overwhelmed again. Because you know, a lot of people are like, well, where do I start with AI? And most people know the power of it, but they don't know where to get started. We show you exactly where to get started inside the community. Plus you can ask me questions and I personally answer them with video tutorials just to help you as much as I can. And then also you can jump a weekly coaching course and that sort of thing. So let's say for example, I am spending a lot of time with ads. I would say, okay, can you add a section inside my tool or can you create a new dashboard for me where you automate ads, right? And then from there, it's going to create it and you have to test it out, okay? So you see how I talked to it in plain English and I just tell it exactly what I want to build? That's what you should do too. So with that, let's get straight into this and I'll show you an example. So if we go to a new chat here and then we're gonna delete this section, we're gonna open up a new folder. Now you don't have to work in a project so you can click on don't work in projects if you want to and it's a new chat to use this. By the way, as well, you might say, okay, do you use chat GPT or do you use codex because there's two different versions of this inside the app. And if you're wondering how to get access to the app, you can just go into Google here and you just get this one, right? So you just download this, this is the app. So once you've done that, you can get it on both mobile and you can get it on desktop as well. Once you've done that, you download it, you open up, you log in and it's gonna look something like this. If you've already got access to chat GPT, but you don't have access to GPT 5.6 and you're on the right plan, just go to the check for updates here, update it, relaunch it and then you should get access. And then from here, you can change the model. And now, once you're inside chat GPT like so, you can switch between chat GPT over here or codex. For me personally, I would stick to codex because I think that if you are using chat GPT directly, like for business stuff and for actual building stuff, like it's much better to use the codex. I just feel like it's much more controlled than that sort of thing. So this is for creating, learning, exploring, this is for actually building stuff. And today, we want to build anything, right? We want to build an automated infant, which is what we're focused on. So we're gonna use the codex section. Now inside the models over here, you can see that we can switch between different models. So if you tone this down, it will use less tokens, which means you will get more out of your subscription without facing limits. However, the lower you put it down, the less powerful the model is gonna be. And if you want to use GPT six, which is the newest version, you would switch up here. Now, you can choose between astralight. You've also got medium and you have extra high. For me personally, I like stick to medium. That's my favor. So we click on new chat, then we're gonna select a new project. And let's say, for example, you just wanted to build an app. Maybe you feel like I don't know, fitness tracking or something like that. So we can put it in a project name like so, maybe you spend a lot of time on this. But everyone like, for building stuff with this, it could be personal or it could be for work, right? So for example, for me personally, my train moiter, I like train moiter, I could create an app for that, right? If I spend a lot of time playing guitar, which I used to, but not so much since AI came up, then I could create an app before learning and playing guitar. Whatever you perfect. My main point here is I, it doesn't have to be workload, it doesn't have to be business-related. Although you'll probably get the highest outputs in terms of your time versus outputs with something business-related. But let's just create a random app here. So we're gonna click on fitness tracking, click create project. And then from here, we can choose between local, new local work tree, cloud. And you can also see here that we can check our usage. And then we've got this option here, which is branches. You don't need to worry about that too much. You don't want to overcomplicate it. The main thing here is like, this is very, very useful to know at this point. So if you're running a new project and you want to share it with people, you're better off doing that in the cloud. And you should probably understand that before you start building anything, before you start prompting it here. So for me, for example, I have automations running in the cloud for SEO. So these cloud-based automations, they create content on my website. I don't need my computer open. I don't need chatchy pretty open because it's running in the cloud. It's not on my local computer, fantastic. 'Cause that can just run 24/7 as a routine. And we'll come onto routines later. I actually add down the list too. My point here is that if you're gonna shut down your laptop, you need something to run in the background, choose cloud. If you prefer something running locally, and it's just you who's gonna be accessing this project, then you can go to local, makes sense, right? Makes sense. So from here, we'll click on local and then we're gonna say, okay, create a beautiful fitness tracking that is designed for mobile devices and make it super useful in action more. All right, so we're gonna start creating that app as you can see here. Now you'll notice as soon as I press enter there, the chat just disappeared, right? Where did you go to press enter? It's not there. It's not obviously here on the left-hand side. Where did it go? This is something that puzzled me the first time I used it. So I wanna explain that quickly here. So what you'll see on the left-hand side, is you have to scroll down. So we have our pinned projects at the top here. Would you agree? And then as we scroll down here, we've got our other projects. So pinned means that we keep it at the top. So for example, if we're looking at that fitness project that we're working on right now, the fitness tracking one, we're gonna right-click that and we're gonna click on pinned. Why do we do that? Because it keeps it at the top and it keeps it easy to track. Otherwise, you don't wanna be scrolling down here every single time. And if you have a lot of projects that you're working on, this is something that's gonna save you time and keep everything much more organized. So it's important to understand that. So if you go to this section here, you can see now that we have the fitness tracking system being built, fantastic. Now that may take a while. Something to understand here is like, sometimes you're gonna build a project. I have noticed that Astra 6 is pretty fast, but it might take five, 10 minutes to build out. So what do we do in the meantime? We start working on the next thing. So you can build out multiple projects at the same time, which is what I'm gonna show you right now. So we've covered how to build an app. And again, you just prompt it in plain English. If you're coding something out, you don't need to be a coder to use this. If you don't understand how to prompt, it doesn't matter at this point. You literally just describe what you want and explain it. The only job that we really have as humans now when it comes to using AI is being able to clearly define what we need and want. And understanding the biggest opportunity in terms of where we can get the most time back. Most people don't have the clarity of thinking to do that. But if you're watching this is a very powerful skill that would highly recommend you develop because it may be the most important skill as we go into the future. So as we have a look at this, let me show you another example. So here we have a website designer, right? So if we open this up, we can see that we have a website being designed over here. So we're just gonna log in. You'll notice that this has been hosted on a chatchivity.site. URL. And you can see the page has been built right there. So if you wanna build websites like this, again, you just type in what you want. So as you said, can you redesign this website? So it looks amazing. And then it redesigned our whole existing website. So if you wanna build a website with AI, pretty simple. You just give it any information or examples of what you like and then you go over to it and you say, "Hey, create this for me." If we go over to. this video section here and let me just take these off as we go along. So if talk about SAS tool design, eth and we've talked about apps as well. Let's talk about website design. This is pretty important to understand. So from what I've seen so far, when it comes to designing a website with chat, it is not particularly great straight out the box. So if you go to chat GPT and let's say we're creating a new website like so. If you just say you create a website for this or do this, etc. doesn't look particularly nice. What you want to do instead is either train it on a skill that helps you create websites or you can actually just give it a screenshot of an example that you really like. So let's say for example, and we'll take a screenshot here. Let's say for example I really like this design for a website. I can take a screenshot of that, then we can go over here and we can say can you create a beautiful website like this for my AI community. Make it clean, follow the same principles, make it super useful and interesting. And so when you give it examples of what you like, it's much more likely to give you something that's closer to what you had in mind. If you do not give it examples of what you like and you don't give it like a screenshot or anything like that of how you want the design to look like, it will probably turn out very bland and it won't be that interesting at all. So for example, if we go back to this page here, to me personally, it looks okay but it's not that interesting, right? It kind of looks like a chat GPT website straight out of box. I've seen Codex follow that same design over and over again. The other thing that I would recommend here is that you can go over to some way like GitHub and you can get like 3D website design projects as well. You can take a look at GitHub, GitHub is basically, it's kind of like a website where you can get ideas for and you can get like these kind of add-ons that you can quickly install to your AI agents so that they give you better outputs. So what that means essentially is that when we are doing website design, we can give it a skill from somewhere like GitHub and train our agents on exactly how to design a beautiful website. So if we have a look over here, we've got a bunch of website design ideas, design systems, here's an example, let's see what else we've got here, page designer as well, that looks like an interesting one. But basically you can get these skills that teach your AI agents how to design. There's also another one called UI/UX, ProMax, which is another website design skill and you can just type this into Google and then you'll find it on GitHub like so. And you can actually go to Codex like so and you can say hey install this skill for website design. So if we go through a new chat here, what type of a new project, I'll call this UI/UX ProMax, paste in the link to the GitHub with the skill and say use this skill to create a beautiful website for the AI profit border. And then obviously when you're creating a website or info like this, you want to train it on information about you and your context. So if we look at this example skill, we would inside the prompt normally give it information about us, what we do, how we're working on things, what we're working on etc. There we go. So that's website designed. There's two different options for that lack of mention. You can give it a skill like UI/UX ProMax to make better beautiful websites or you can actually give it an example of a screenshot and say hey go off and create this. Now what you're also going to notice here is that we now have three projects being done in the background. So we can check on these. If you see this loading icon here inside codex, you can see that we can check through these different systems here and we can see what's being built and where we're up to. So for example if we click on the UI/UX ProMax project, we've got an update and it says it's creating the website. If we go over to the AI community section here, you can see that it is now working on building out the other website. And if we go to the mobile fitness tracker here, boom shackle hacker, we now have the website ready to preview as we can see right here. Now what you'll notice is we have a preview on the right hand side like so and we can flip through we can test it and that sort of thing and we can actually open that up in Chrome as well. Now some people at this stage might be saying as well okay well that's great that you've got the mobile app but how do you for example submit it to the app store? What you would do is you would say okay I want to submit this to the iOS app store can you get it ready to export? So that will be number one and then number two if you wanted to for example create an app that you're going to issue publicly, you need to make sure that people can log in. So for example if we look at this whole app here, nobody can log in. The data isn't stored anywhere. So like if if I'm using this and then someone else is using it, well how do they log in and they store their data? So what you want to use is something called super base. Super base is another free plug in. You can give to your AI agents like codex and that can then plug in super base as a database for people to log in to our apps. So what we can do here is we can say okay make sure that people can log in using super base so that you can store data and you can have a database of data for each person who's using the app. And then also the final thing that you'll notice here is like the colors look a bit bland and a bit boring. So anytime you have feedback if you test something it doesn't work or for example you look at something like I'm not really into that design that's not my style. You just give that feedback to the AI agent and you iterate from there. Bear in mind like you can get the 80/20 very very quickly. We've got the 80/20 of an app right here. Like that's good to go. The problem is that we need to customize it so that it doesn't look like everything else. It doesn't look mediocre that is actually used for a set room. And so how do you do that? You just give feedback to your AI agent over here and then it begins to code out the projects on the right hand side. So for example if we say okay can you start improving the UI making it a bit more colorful, making it much more interesting, entertaining to use et cetera, gamify it as well. And see see how we give the feedback to their agent. We make sure it has a database for using that. And then also we're going to give it the details for exporting it to iOS and Xcode and everything else. Pretty simple and easy. One of the questions we just got here actually is, you know, who do you think is winning the AI race right now? Is it GP6 or Fable 5.1? I think in terms of actual intelligence in terms of the best AI to use right now, undeniably, it is GP6 astra. I tested out Fable 5.1, it's okay. But honestly for me I think Claude needs step up a lot at this point like GP6 astra is way way better. So you can see here that it's beginning to now come up with some ideas and it's going to give us options. So for example we're going to just give some options here and then it can start using that to fill out the rest of the details. Now what you see is the page for the community. So you can see here for example we give it details of the page you'd want it to create and it's created over here. And so you see like when you when you give your AI agents examples of the UI or the design that you want to do whether that's with screenshot or whenever it will look 10 times nicer and it just won't look like you generic sort of AI fluff. You know see here it's asking me to log in to Superbase which you're doing in the background but this is a page that's created for the website that we asked it to build. Pretty nice and easy. You know and it's it's created that community page right here. It's pretty cool. So we've built out a website we've built out another website over here with the UI UX Pro Max system. So if we open this up this is actually hosted locally and you can see that basically chatGBT has its own browser over here but what you can also do is you can load that up locally inside Chrome or whatever you prefer to use. So if we go to Chrome here and we typed the local addressing you can see that we have the page ready to go. So there's found the details of the page that you want to build what we want to add inside there and that just looks a lot nicer as well. So it can easily build out designs and that sort of thing. The first thing I'll say here is like it's still it's still not customized exactly how I want it but again that's where the feedback stage comes in and you'll notice that for any sort of project that you build with chatGBT whether that's a video, a game, a website, a tool. There's basically three stages. You describe what you want then you test it and then you iterate with more feedback and you just loop around. So some people watch this and be like oh you know that's not the best design in the world Julian and it's like yeah I know I know you need to do a lot of feedback you need to give a lot of feedback to AI agents just to test it go through it see what you think about it and customize it as out how you want. So if we go back to for example that tool that I built earlier, the Aging OS, which again is available inside the powerful volume. This tool is amazing. But it took me day, I mean months at this point of like just more feedback every single day, right? Create something, build something, test it, give it more feedback, loop back around, build a new feature and test a loop back around. And so like if you see anything today that you're like, oh, I could do better than that. Everyone could do better than that. You're just going to take a bit more feedback and a bit more time. Just something to bear in mind there. So we've talked about website design. What else we got here? Images. So chat, GPT can generate images. Now it's using something called chat, GPT image 2.0. So when you're using Astra, Astra 6, you want to think of it as the brain. So it is the brain. But image 2.0 is this image generation skill. And so when chat, GPT generates an image, it's using chat, GPT image 2.0. So Astra is not actually creating the image, but it can orchestrate the skill to build the image for you. So if we go into a new chat here, we don't need a project for this. So I can just say, OK, create an image of a cat eating pancakes. Right. And then you can see here that it will start building out the image. Now at this point, you might be thinking, OK, that's great. But why would I just want to generate one single image? And I would say you've got a great point there. So when you're generating an image, you don't just stick with the image. You can use it for social media. You could create an image for ads. You could create thumbnails, for example. You could create an image and then use it on your website design. You could take an image and then get Astra to turn that into a game. So my point here is like, image is very powerful because you can animate those. You can take them and transform them into websites, apps, games, tools. It can make all of your pages look a lot nicer. So if we have a look at this, for example, this image right here, we can now go back inside the chat and let's have a look where we are now. Again, you see how we have to scroll all the way down to this recent chat. We can comment on here. We can also remove the background. We can remove it or resize it, as you can see. So what we can do here is we can say, okay, can you, using your skill for videos, emotion, could you turn this into animated video, just like a quick 10-second promo video? Can you also use this image to create like a 3D racer game or something like that? So I'm going to show you two different examples where basically you can get an image like this and then you can turn that into an actual project. So you can see that's now beginning to work on both of these. Now in the meantime, we've talked about images. That's where we're going to come onto videos. So video is pretty cool with chativity, astra, because astra number one is fast and number two is very, very good at coding. And the way that AI can create videos is with code. It's using the skill that basically creates HTML. And with HTML, it can generate a beautiful video. So for example here, we have the fitness track of being built. We also have the cat pancake images you can see. But if you go back to the video section here, you might go on and okay, how did you create this promo video? It looks pretty nice. It's pretty simple and easy. So what you can do is you can use something called remotion. Now, remotion is a skill you can install to your AI agents to basically create videos. And it's a free open source skill, which means that you can give it to any AI agent, you can create videos energetically, and you can create these awesome animations just in plain text. So what you would do is you would grab the GitHub link of remotion, then you're going to go over to chatchy beauty or whatever agent you prefer to use. And then you can say, okay, use this and create an amazing promo video for whatever, right? It could be like, okay, create an amazing ad, create an amazing promo video, create an amazing presentation, create an animation. You see my point like anything that can be animated, you can do this with this system. One thing that I actually really like to do, and this is something I built as a custom workflow, but I'll show you it for inspiration. So inside our agentik operating system, we have a video agent and it uses remotion to create amazing videos like you can see right here. And if we go back to our videos over here, you can see some of the other videos that we've created. And these all have like audio in the background and that sort of thing. And it's pretty crazy. Like some of these videos, they are quite long videos. So let me show you an example here. So this one is a five minute long video. And what it's done is it's used a combination of another tool for generating the B-roll, then it's created a voiceover and a script for the video. And it's put it all together and added it. And literally all we do is we go inside here and we type in the type of video we want and it can generate it and save it into our workspace. Here's another example. So this video, you can see an AI avatar down here. And you can see that he's talking on the screen. And it's creating. Right. And so my point here is like, it doesn't even just have to be for a basic animated video. Like you can create something pretty unbelievable. Just just by combining remotion and other APIs. So for example, like for AI avatars, you might use API like Hagen. You can grab the API key from Hagen. Then you plug that into the astro conversation you have for the video. And you say using the combination of this Hagen API and your skill for videos, remotion, create animated, edited avatar video. And that's basically what we did with this video right here. It's pretty cool. Now, if you come back to our project with the app, I want to show you an example of logging into the app. So this is a mobile app that we built. We just need to sign in to Superbase. And then it will allow us to to log in and create a login for that. So you can see here that we can now log in to our account on that habit tracking app. And it will save our data using Superbase. And I don't need to edit it. I don't need to add any of that in. Astro just logs in for me. And once I've logged in, then with Superbase, we can add the login page right here. So if you come back to that example with the image, you see how it's created this image for the video. And it's generated the full animated video right here. And then it's created this ready set pancakes animated promo using the image inside the video itself. And that's just a combination of emotion as a skill. And then the image generation skill, which is with chat GPT. That's another big benefit to using something like a GPT versus Claude. Claude can't generate images. You have to give it an API. So you have to give it access to image generation API, like for example, chat GPT, which you put up, which is the best one. So we've chat GPT inside your assistant subscription. You get access to this as well, which is a massive benefit personally. So you can see here, it's now created in the game as well, which we can open up later. That's essentially it for videos. Now let's talk about computers. So we're at the point now. And honestly, before Astro came out, this was not particularly great. But now we're at the point with AI where it can control our computer, you know, it can control pretty much anything on our computer and do anything that a human would do. Now it's not perfect. It still struggles in part. So I want to be 100% transparent with you. But it's way better than anything else I've seen. And it's pretty impressive so far. So for example, if we go over here, we have chat GPT and the Chrome extension setup. If you want to find the Chrome extension, I'm just going to explain this because some people struggle and they put this in the comments sometimes. So if you're looking for the Chrome extension for chat GPT, you would just go to Google and type in Chrome extension chat GPT. So anything that you're trying to find today, whether that's like remotions, skills, whatever, you can just go to Google and ask it for it. So for example, here you can see we've got chat GPT. We just have to add that to Chrome. Once we've added it to Chrome, we log in. And now you can see we've got chat GPT over here. Now for me personally, the Chrome extension is not that useful. The only thing that I actually use it for is computer use. So for example, if we go over to chat GPT here, and by the way, it's actually created the game based on the image we generated earlier. Doesn't look that similar to the original image, but I mean, look at the game itself. Pretty nice. 3D game nicely edited. It just created it in the background once I was talking to you as well. So you get the idea like it's really, really good for generating and creating games and all sorts of cool stuff. Anyway, if you go to a new chat here, and then we're going to open up a new project, we'll say, okay, we'll call this school post, for example, like this. And then I'm going to go over to my school group, we'll copy that and paste that link in, just as a reference point. We'll say to Chad Chippity, "Hey, can you go to the page above using Compute Use?" And then what I want you to do is make a post about Chad Chippity Astra and how we have loads of great new training inside the AR Profit Boardroom on exactly how to use it. And also a bunch of free bonuses inside that too. And then I'll just give it a link to the AR Profit Boardroom and I'll say I had urgency, make it interesting, throw grade, reading level, down Kennedy style, marked it post and create a new post. So this is an example, right? We're going to hit Enter like so, so we can see that is now thinking and it's going to open up the community. So you can see here it's beginning to open up the school community, and so it's connected to our Chrome and we can see it's opening up Chrome right here. Now you saw how fast that is, really, really quick when we're using it. And so if you look, we know the Chad Chippity is using that tab, so it's having a look here and here and it's controlling our browser. Now look how fast this navigates. So this is Chad Chippity right here controlling the screen and navigating across it. And now it's just put in the title, right? So it's added the details here and then it should put in the post. And the crazy thing about this is you can navigate the buttons really quickly, you can scroll around the page, this isn't AI controlling the computer and it's unbelievably fast. But also if you got a virtual assistant to do this and you gave them a race, like how long would it take to create a page like this, probably ages. So you can see it's now created the post, right? So you can see it's created the marketing post for us, it's added a couple of links. The other thing that would say is that it's actually using information from the page originally and then use that inside the post if that makes sense. So it's not just like generic AI fluff, like it's actually personalized the post itself to the information it saw on the sales page for the air profile water, which is fantastic. Also, I like the way it's written, like it's easy to understand, it's very clear, it's beautifully formatted, it's added urgency, I mean, and it did it in first time around whilst I was talking to you. Now, I think at this point, if you've got a basic task, like for example, posting and you need it to come up with a creative in terms of the writing and what's going to be in the headline, and also post it for you, we're at the point now where I think you can get our stress to do most of that for you. And you could even have that running as a schedule in the background which you'll come on to in a second, but yeah, really powerful. And then it's said, published, you know, only took 90 seconds. Imagine how long it would, if you've ever hired a virtual assistant or a human to do stuff like this for you, they might take like 10, 20 minutes, 30 minutes thinking about this. This has cut that down, that time window down into 90 seconds. That's how powerful GPT-6 asterisks. And it wrote it better than most humans possibly could, it's personalized it, there's going to be no mistakes in terms of the spelling or the punctuation inside the post because asterisks are very good at understanding graphical rules, and it's published and it's ready to go. Okay, just happening in the background, so so fast and easy and smooth. And I can tell you now if you got, for example, like Fable 5 to do the same thing, it would struggle. It doesn't do it quite as well. It can do the same job, but it'll be much slower. And I would say the content itself in terms of the post, the headline, everything else, just wouldn't be as nicely designed as this. So this is way nicer. All right. So pretty awesome, and as you keep going along now, we've covered computers, we've covered apps, stats, tool design, website design, images, video, computers, right, we've covered a lot already, which fantastic. I want to show you an example of the game design that we did earlier. So I've shown you a couple of games that we built with them already. So we created the 3D race, a game with the cat. If you're not sure about that, it's over here. This is the game that we just created in the background. So this is the game, and it's even got like controls, which is pretty fun as well. So we can pause this and we can play around with this and whatever. But then also, if we go to the app, we'll build the app. We created the video earlier today, as we mentioned. Now this is where it gets quite interesting. So we can also create cloud routines, schedule tasks that run in the background. So what do I mean by that? Well, let's say, for example, you add a website, right? So if I go to the aiprofitboardroom.com, this is my website, AI. And we do a lot of SEO for this website. Now the great thing about AI, let me show you this is the results we've had with this website. The great thing about AI is like, you can definitely get your results for SEO. So for example, if you look at this website, we've grown it from my zero to four hundred and twenty three clicks a day using AI SEO. And it's fully automated. I can't even log into this website, which is amazing. And by the way, if you're on a free SEO strategy session, link in the comments description and we can give you a custom tailored game plan for this or stuff. But my point here is that you can get AI to automate your SEO for you. However, you still have to prompt the AI to do this daily. You still have to come up with a keyword. You still have to check that the work was good. And you still have to have your computer on and make sure that you're remembered to create that content on a day to day basis, which, you know, in the long run of things, that might take you five or ten minutes a day, which is like, you know, 150 to 300 minutes per month. That's like nearly six hours a month. So well, five hours a month, sorry. But you get my point, like, why would you spend five hours a month doing something that you could automate that would save you more time and that's where schedule tasks come into play. So schedule tasks basically allow you to run routines with AI so that something can run in the background without you, which means that it's fully automated and you never need to think about it again. So if we have a look at this, for example, I said to chat GPT, can you set up a cloud routine that runs without having the codex app open in cloud so you can generate content for my websites. And then he asked me, like, which website should I cover, you should cover my five websites, say, oh, no, Netlify. And it says where does the other computer save the Netlify website and said Netlify? And then it created a cloud routine as you can see right here. So this is a routine that happens in the cloud. Now, you could have a local routines, well, but I don't like to have local routines running simply because you have to have your computer on and you have to have the codex app open. It's kind of a limitation because, you know, not everyone's going to have a computer open all the time. Well, sometimes you ask them to close the app, not great. What you can do with this instead is that you have a cloud running routine that just runs in the background. I totally forgot I set up. I didn't even know how to access it for the first couple of days. And so this is amazing for AI research for SEO projects. For example, something that's recurring on a weekly basis. It would be amazing, for example, for anything that can run in the cloud. Now, the one thing that I would say here is that it's not good for computers. So if it's late, for example, posting to X or something like that, probably wouldn't recommend it. If it was something more API related, like, for example, creating SEO content and then publishing it via Netlify via the API or the connector, that would be great. That's what it's designed for. So have a think about all the recurring tasks that you work on day today that you could easily automate with this system, because when you think about it, there's so much, there's no limitations to that, I think. So it can easily create cloud routines for you. That runs with codex closed. It's got my SEO skills embedded into creating it. It runs on the daily routines. You can see here it's scheduled for 7 a.m. each day, and then it can create five articles for website and it's automatically configured to publish and it's ready to go. Now if anything breaks on that routine, the first time you set it up, like creating the apps, like creating the website, like creating the videos, you have to give it feedback in the first few days, but the more feedback you give it, the better it becomes over time. It's definitely something to pay attention to right now. So that's something else that's really cool. And then also we have something called plugins and connect it. So if you go to the top left over here, we can make GPT-6 astra infinitely more powerful because we can connect it to other apps that allow us to take the power of astra and not have it confined to this tiny little app. So just like the cloud routines connect to Netlify, we can connect astra to some of the biggest apps in the world to do recurring work for us. So for example, we can read and manage your Gmail and if you want to set up any of these, you just go to plugins in the top left inside codex, right? So you go to codex, plugins, and then you can see here that we can switch between personal and public and then we could, for example, set up Gmail here. that we can. actually try Gmail, like you can see, and then we can tag it in this workflow. So if we type in app, and then Gmail, we can tag that in, and it's ready to go. And then you could set up a new routine, like we were talking about before, where it analyzes or answers your emails for you, where it drafts all the responses to your emails for you, whatever you prefer. But my point here is that you can connect the stabs, they can run on the schedule tasks, and then you can basically automate a lot of the, you know, a lot of the work that you're doing today today. Save so much time, definitely worth doing. The other thing that you'll notice here is that we've got this option for full access. Now, if you want to limit the amount of access that your AI has, especially if you're coding on a project, you've already developed a lot of time on. Or for example, if you have certain files on your computer, the don't want chat you be able to touch, then you can switch between three different levels of approval. And so for example, you can ask for approval ways, like so, you can say, approve for me, or you can give it full access. Full access means that it's got unrestricted access, you can edit files, you can delete stuff, etc. You can also have a proof of me. So if it identifies anything like, for example, delete a file, then you can switch to approve for me. Or you can switch to ask for approval and ask for approval. It's kind of slow. It's kind of annoying because you have to like approve everything that does. But at the same time, if you want to play it more safe, that's probably an option to you. We can also check our schedule tasks over here. So we can create new schedule tasks. These are running normally locally. So you can just say, okay, let's set up a schedule task. Maybe you check their blaze AI news headlines every day. So for that. And then you'll see inside the explore section here, we have options to customize. So for example, for me, I don't connect this to pull to get help. So I'm probably going to uncheck the pull requests. And then we can click on explore. And then size, like so. And we can see all of the websites we've built. So you know, we were just building websites a second ago, Lisa stored. So you can come back to them here. Now, also if you're building out like this, you can have multiple different chats inside the same project where you work on multiple different things. So for example, we could have one chat for UI. We could have one chat for adding new plugins. We could have one chat for setting up. So it's ready to go with X code. And so you can have inside the same project parallel conversations going on editing different angles of the project. It could be like part of it. It's back end part of it. It's front end so that your app is much more useful over time. So yeah, there's just there's so many things you can do with this. Now you can use it inside chat, you be on work. Like so. I just I just don't feel like it's as useful there. But what I do like to do is I have this agente, got brain system. And any sort of custom workflows I'm going to come back to. I like to build into an agente got brain system into an agent OS mission control. If you're wondering how to set up a mission control dashboard like this for all AI agents, where for example, you have called Hermes Antigravity Codex or built into one dashboard that you can control. You would literally same processes I've talked about earlier, you would go into your AI agent and you would say, Hey, build out a mission control dashboard for me. Be build out a agent operating system for me that I can control. And if you want to get my system that's ready to go with all this built in, you can get it inside the air profile bottom, just go to the classroom over here and then go to new daily updates and you can find our agent OS system that we keep updated. It's got a full video tutorial and there's a files world to install. So the cool thing about this is that we can use GPT 6 Astra inside our agente operating system. And we don't have to pay for the API and fly that because we've already got it included as part of our subscription. And then we can use it alongside Claude and Hermes Antigravity. And I think that's the most powerful way to use AI. And also because for example, I mean, this is a big problem that I think most people are facing with, right? Is I imagine, for example, you have probably some people watching this would be like, Yeah, this is me. Imagine for example, you have chat GPT and a subscription. Then you have for example, Claude and then you have for example, open Claude and then you have hermys and then maybe you have, for example, custom workflows and dashboards and other tabs. Everything gets very messy. Everything is all like all over one place. And the thing is like then you're switching between tabs, you're all these different apps, open, they're slowing you down, they're confusing you, you kind of forget where stuff is and it's all a bit overwhelming. That's the problem that agente operating system solves. And what we've done is we've built in codex into it. So you've got a GPT 6 astra here and even if you don't use codex, I don't know why you'd watch a course like this about GPT 6 astra if you didn't have codex, but even if you didn't have codex, well, you can switch to Omniru or HY3, which are both free AR models you can plug into this system as well. And then over here we have hermys. So what you'll find here is that we have GPT 6 astra built into hermys and I absolutely love this because what this means is that you have an agente canis, which is hermys itself, then you have the brain that plugs into it, for example, like GPT 6 astra. And then you have all of your different AIs plugged into this swap. So for example, we have Gemma 4 over here. We have for example, QAN 3.7 over here. We have a custom-trained version of deepseek over here and we have like local models plugged into hermys as well. And I can just click on the drop-down switch between all of them. That is so useful. And then also what we have is we have different versions of this SIGENTIK operating system depending on the workflow I'm trying to use. So if I need to do research, you know, I've been mentioning about doing schedule tasks with research. Well, we can go into hermys with GPT 6 astra and that just runs on a 24-hour loop and it gives us the latest news in our industry. So you'll see here, for example, we've got GPT 6 astra. We've got the latest news here. It's inside one beautiful workflow. It's using GPT 6 astra as a brain and then hermys agent as a harness. And this is like a custom beautiful workflow inside one dashboard where again, it's just running on a loop on a schedule on a routine and I don't need to be there to check it out. I can also hear the brief here so I can listen back to the brief file one, two. We can consult the Oracle if we want to refresh hermys Oracle. We can actually take the ideas. So for example, for GPT 6 astra is trending, then we can actually click on draft this draft post for Twitter or we can click on publish to WordPress and that will actually publish an SEO article directly to WordPress on that particular topic. And we can see previous news headlines from the days before. So it's really easy to stay up to date and see how the industry changed. Not only that, but something else that you'll find with AR models in general and I see this all the time is that people struggle with context. Right. So for example, if a new model like astra comes out or Claude comes out or for example, let's say, I don't know, GLM 6 comes out tomorrow. These models are probably not ones you're using day to day. The brand new models have just dropped and sometimes that means you need to switch over to that new model. Now from with that is it doesn't have any training or personalization or understanding what you've done previously. So how do you fix that? The way that you fix that is you have a second brain that plugs into all of your AI agents and not just that, but your AI agents actually update it for your automatically and then it becomes a positive feedback loop where basically your AI agents update your memories for you based on what you're doing. So you can see, for example, here we're building out that app with codex and it's plugged it into our memory system. Amazing. And then we have our memory galaxy here visualized and we can see all of our memories inside one place super useful. So what this means essentially is that if we do something with GP6 astra, well, that is updated inside our memory inside our second brain and all of our AI agents see that and then they understand what we've been working on today as well. That is amazing. Super useful. It saves so much time. Now if you're wondering how to do that, how to plug in system like this, you want to get obsidian. Obsidian is a free memory plug in. It's open source and you can get your AI agents to update it for you. So this whole database of memories, all these notes, they're just simple markdown files. I don't update it myself. I've never created any of these files myself. But what it essentially does is it updates everything for me. So all of the latest files inside here if we go to, for example, let's have a look here. Yeah. So if we have a look at this example, codex has updated obsidian with the details of the pancake eating game that we just built a second ago. And it did automatically whilst I was talking to you. So it's running in the background. It's stating all my agents what we worked on recently. And then our AI agents can pull that context from this system. So it's free. It's open source. The agents manage it for you. So it saves you a lot of time. And then if a new agent comes out tomorrow, let's say, for example, everyone stops using Hermes in there and something more useful comes out? No problem mate. what's going to happen is we can just take this memory and we can plug it into whatever comes next. And then it's always improving. And it doesn't matter what model comes out next. Well, you know, GPT 10 comes out. No problem. It's going to have a full understanding of me, my life and everything I worked on from day one. And that is absolutely amazing. So that's where I'd recommend. And I think there's levels to this. So level number one is that you just create in one basic project. Level number two is that you're automating a few different things from running schedules. And level number three, the best and most liberating model I think is, is basically the agent operating system and having codex and Hermes and FNL's plugged into one place. And what does that have custom workflows like, for example, that video agent that we built before? So we talked about Hermes. We talked about routines. We talked about the agent operating system. And we talked about where to start, which is basically you want to start where you spend your time. And that's basically it. That's the whole system. We've talked about pretty much how to automate everything. You know, how to get access examples, automations. We'll build a few things today. Apps, SaaS tools, website design, images, videos, computer use. We talked about game design with Blender as well. Brutines, Hermes, Agent OS, and also the plug-in systems as well. So that is basically it. That is the full GPT-6 Astra AI for course one hour on how to build an automate. Anything, if you want more training on this sort of stuff, or if you want to connect with me personally, you can do that inside the AI Profit Boardroom. Link in the comments description or just go to the aiprofable.com. Inside the community, you can ask questions to get help and support whenever you want to. So you can see, for example, over here, we have a four community where you can ask questions. And I personally answer these questions every day. Plus, if you have like a question that you struggle with, for example, like this, you can see that I will create a personalized video tutorial for you and give you my notes breaking down exactly how to handle that situation. So the great thing about this is you get everything you need to win. Plus, we all help each other out. We will share wins. We will help each other out and give each other answers and try and give each other advice that's useful. And that's fantastic. Also, of course, stuff like, for example, we actually just dropped a new course inside the AI Profit Boardroom. And I do a weekly update. So you can see this video update here where basically I help you as much as I can. And I just break down, you know, what's useful this week and what's actually helping. Inside the classroom, you can get access to all of our best trainings. So if you're a complete beginner, you can start with the beginner to expert course in 2026. And we break down step by step how to get more leads, how to get more customers, how to grow your business with AI. We've also got a full section where you can start and learn anything you want. Inside the new daily update section, we have a advanced tutorial. So if you're more than a beginner, if you're if you're advanced and that sort of thing, you can check out our advanced trainings. And then if you have any questions or if you need personal advice, you can post inside the community too. We'll say I have all of our newest and latest trainings. As you can see right here, so you can access to all of these new trainings and courses, which is pretty amazing. And we have four weekly coaching calls where you can get help and support in real time. So the great thing about this is the, you can jump on those calls, you can ask questions, you can share your screen, you can meet other members, you can see what they're working on. It's just an amazing community where we all learn and we grow together. Inside the map, you can meet people in your local area who are building out with AI agents like you. And this is all available inside the Air Profit Boardroom. Link in the comments description will go to theairprofitboard.com. Thanks for watching and I'll see you on the next one. Cheers, bye-bye.