Master 80% of Antigravity in 25 minutes (You’ll beat Everyone)

Google's new AI coding tool, anti-gravity, has completely changed the way that we're now able to build applications and softwares just by using a single prompt. But the problem is that most tutorials that you see out there on

YouTube get way too technical and completely skip the basics of how it all works. And so in this video, I'm going

works. And so in this video, I'm going to show you exactly what anti-gravity is and how it actually works. I'm going to reveal 10 hidden features that separate beginners from the pros. And finally,

I'm going to build a complete application right here in front of you so you can see the entire process step by step. And in case you don't know who

by step. And in case you don't know who I am, my name is Mikuel and I've scaled my own AI agency to six figures in 14 months, helping over 50 businesses to implement AI and automations and tow

over 20,000 people in the process. All

starting with zero technical knowledge.

With that being said, let's dive in. All

right, so this here is a diagram that shows you exactly how anti-gravity actually works from top to bottom. The

first thing is you. So you are an individual, you're a human that do two different things. The first thing is

different things. The first thing is describe the task, which means that you describe to the actual agent what it is that you want. Whether you want to fix a bug, whether you want to build a to-do app, whether you want to build a finance dashboard. And the second thing that you

dashboard. And the second thing that you do is edit code, which means that you review the work. Okay? You don't

actually have to touch the code, but you check if AI did the right thing and you approve or ask for changes. Now, the

good thing about this is that you literally don't have to touch the code.

All you have to do is look at the output that it gave you, and if there's something wrong, then you tell the AI, "Hey, can you edit this part?" Now

everything that's within this exact square is the app itself. So everything

here is what anti-gravity gives you natively. And so you as a human you

natively. And so you as a human you describe the task to the agent manager.

Now the agent manager think of it as mission control. It's where you're going

mission control. It's where you're going to type your task or request to the actual AI. It's where you're going to

actual AI. It's where you're going to see your conversation history. So the

actual inbox and it's where you watch what the agent is doing right here. So

this is the manager for all the agents.

Then on the other side we have the editor. This is the VS Code looking

editor. This is the VS Code looking part, right? A normal code editor where

part, right? A normal code editor where you can see your project files, right?

You can actually edit the code manually if you want and you can run terminal commands right here. But this is essentially the main hub where you can add all the files. You can uh you can look at the code that it's doing and all

the different you know areas that it's actually working on. So again you as a human you have to describe the task to the agent manager and you have the edit code that is going to be in the editor.

And a good thing is that you can switch between the agent manager and the editor at all times. These are just two different views of the same app. Then

you ask yourself who is actually making the software because VIP coding is just you putting a prompt and getting something out of it, getting a software.

So the way that it works is that when you describe the task to the agent manager, this sends the request to the actual agent. And this right here is the

actual agent. And this right here is the AI worker. When you give a task, the

AI worker. When you give a task, the agent plans so it figures out what steps to take. It codes so it writes the

to take. It codes so it writes the actual code so you don't have to. And

then it runs, which means that it executes terminal commands. So it

starts, you know, taking action on what it needs to do. And finally, it also tests. So, it opens the browser to check

tests. So, it opens the browser to check if it actually works, which is a beautiful thing about anti-gravity because it goes in your browser and acts as if you know you're going there, which means that it tests on your behalf. And

a good thing about this as well is that the agent is autonomous, which means that it does everything on its own and then it comes back to you for approval.

Then you ask yourself, what does the agent use to actually make the software?

What you use is three different things.

So, the first one is the model. Now,

this is the AI brain. See, this is the intelligence powering the actual agent um that's making the appro software. So

you can use Gemini 3 Pro and which is Google's then you can use Claude which is anthropics and GBT which is open AIS.

So different brains have the same agent.

So you can just choose which one works better for you. Now typically Gemini 3 is best for creating you know nice looking front ends. Front ends is just like the way that the software look.

Then we have cloud which is more backend stuff. So more you know writing the

stuff. So more you know writing the actual code and getting sophisticated with that. And then GPT is more for

with that. And then GPT is more for everyday tasks that's more fast and easy. Then we have the browser. Now, the

easy. Then we have the browser. Now, the

browser is a built-in Chrome uh browser, right? The agent can literally open

right? The agent can literally open websites here. It can click buttons. It

websites here. It can click buttons. It

can fill out forms. It can take screenshots. It can record videos of

screenshots. It can record videos of what it's doing, which is exactly one of the big differences between anti-gravity and lovable bolt. And this is how it's going to test your app because it actually uses it like a human would. So,

you don't have to. And then finally, we have the MCP servers. One of the most powerful features of anti-gravity when making apps because this means that you are literally connected to any software that there is out there basically as

long as they have an API. This can be connecting your software to N10, connecting your software to Slack to Google Sheets. And so the agent can do

Google Sheets. And so the agent can do things outside of just your code like post a message, send a message to any to maybe trigger an automation uh or pull in any single data that it wants. And

I'm going to show you exactly how you can connect any to anti-gravity as well to be able to make these softwares come all together. Now the agent itself when

all together. Now the agent itself when it is creating the software, it does also create artifacts. What this means, and I'll show you exactly, you know, when I actually build it. When I tell the agent, hey, I want a software that

is a finance dashboard. Well, the first thing it does is plan, right? And so it starts making task list. It starts

making a plan. Sometimes when it does test, it takes screenshots. It takes

recordings and those are things that we call artifacts which are all stored in our workspace. Okay, so the full project

our workspace. Okay, so the full project is stored in our workspace. Our

workspace is basically just a folder that you make on your laptop and all the files that the actual agent is using are all going to be stored there. So that

you end up having different folders in your sort of desktop, right? Which all

represents different projects which again are files and also the code that the AI agent built and the files that it uses to make that code. So now that you understand the basics of how it all flows together and connect with each other, let's dive into the actual

platform and show you how it works. All

right, so the first thing you want to do is you want to be able to download the actual software on your MacBook. And so

you can either download for Mac OS here or you can press download here and it will take you to a page where you can download it for Windows, Linux and Mac OS as well. Once you downloaded the software, you will get on a page that looks like this. Now it might seem a bit

complex because it starts showing you folders, manager, repository. It looks a bit techy here and there. But it

actually is very easy to understand. So

there are three different components to it. There's one component here, the left

it. There's one component here, the left hand side. There's a second which is the

hand side. There's a second which is the middle and then the third one which is here. Okay. Now this is what we called

here. Okay. Now this is what we called editor. It's where we edit the [music]

editor. It's where we edit the [music] actual things. All right. So on the left

actual things. All right. So on the left hand side we have the open folder. So

this is where we get to make the folder which is going to be on stored in our MacBook or our computer. Why is because all the artifacts which is all the files, the whole code, all the different things that it uses to make the actual

software will be stored in one place and that's going to be that folder which is you know the workspace. And when you open this folder so let's say we open it up, we create a new folder and we name

it test NA10. We can create this and now I can open it and then I can put yes I trust the authors and right here I can see that we have opened a workspace and

now I can start adding in files right here to the workspace that the agent can actually use when making the software.

Okay now once we have made this in the middle right here I can see the code being built live and on the right hand side you can think of this as the chatbot that you talk to sort of like the AI that you're actually using uh to make the software and so here is where

you actually tell it to action on something. Now right here you can see

something. Now right here you can see that you have the different models. So

this is the brain Gemini 3 Pro uh low flash here and GPT as well. Again Gemini 3 Pro is great when you're building web apps just because it is very good to make it visually appealing while Cloud Opus 4.5

is something that you know you actually use when you're making something that's a bit more complex when you're making software and then finally GBT for you know easier tasks as well. And then on the left hand side you can see that we have two different types of conversation

modes. We have planning and we have

modes. We have planning and we have fast. And what it is in a very easy way

fast. And what it is in a very easy way is that when you tell the agent to do something. So let's say you tell it to

something. So let's say you tell it to make a software. Do you want the agent to plan before executing the tasks which is planning before executing the task?

Well, yes I do. I want to see exactly how it plans throughout the actual way of making a software so that if anything goes wrong, I'm there. Okay. On the

other hand, we have fast. So the agent will execute the task directly. This is

used for simple tasks that can be completed faster. So if you are making a

completed faster. So if you are making a software, you typically want to have it on planning mode to make sure that you're all good. Now, one thing that you do want to do is you want to make sure that the agent when it's making the software, it doesn't have to ask you every single time if this is right or

this is wrong, right? Cuz it can be overloading, especially when you're making something a bit more complex. You

don't want to get questions all the time. And so, the way to change that is

time. And so, the way to change that is you want to go to the settings on the top, you want to go to open anti-gravity user settings, and then right here on agent, you'll be able to see that this will be turned on. So, you want to turn

it off. And you want to make sure that

it off. And you want to make sure that the review policy is turned to always proceed and terminal command auto execution is turned to always proceed.

Now, is this the safest thing to do? No.

But when I'm building software, I don't want to be there. As in I can go get food, I can go get a drink and so on.

And when I come back, I can make sure that it's done without me having to answer every single question and making sure that I'm there for every single step, which is why I put always proceed.

But if you're doing something a bit more sensitive, then you might want to have this to request a view to make sure that the agent always gets your input for every single step. So that right there is the main editor interface. Now on the left hand side you can also see that we have code search, we have source

control, run and debug, remote explorer and extensions. Now code search is when

and extensions. Now code search is when you want to find something within the code. I don't typically use this.

code. I don't typically use this.

Neither are you to be honest, especially when you're not a coder. Uh which is fine. Then we have source control, we

fine. Then we have source control, we have run and debug, and we have remote explorer. Three things that I don't

explorer. Three things that I don't really use. The main one that you're

really use. The main one that you're going to use um out of this three would be run and debug in case you know something is wrong with the code and you want to debug it. Debug just means fixing the code. You can see everything here. And then finally, we have

here. And then finally, we have extensions, which you can think of the app store for anti-gravity, which means that you can get these different apps and connect them to the software to be able to add functionalities. Not

something that we're actually going to use uh when making the software, but something that you could do to add more things over time. And lastly, from the editor settings, we also have the customization in MPC servers. I'm not

going to dive into this in specific. I'm

going to dive into this later on, but customizations is basically when you want to make the agent better. You can

also add more rules and I guess more files that you can just always reference in MCP servers. If you press on here, you can then look for the MPC server.

Now, MCP stands for model context protocol. And what it means is that you

protocol. And what it means is that you now basically have the ability to connect to any software that you want in this list. But also, if you press manage

this list. But also, if you press manage MPC servers, you can also press this button right here, which is view raw configuration, and you can add your own API, which you are going to do when you're connecting any to anti-gravity.

And now that we're done with this, we can press open agent manager. And this

will take us to the agent manager right here. Now, for the sake of this video,

here. Now, for the sake of this video, because I want to keep it simple for now, I'm not going to dive too deep into the actual agents and how they work and actually testing them out. Uh, but if you want a video on this, you can let me know in the comments down below. I'm

happy to do it. But all it is is for every project, every workspace that you have, there's going to be at least one agent working on something. Okay? So, at

least one agent may be working on a software app or maybe working on the interface of the of the software that you're building. But then you can also

you're building. But then you can also run different uh agent in parallel so simultaneously at the same time to be able to have different agents doing different things at the same time when you are making the software and you do

this by starting the conversation which you can here again it looks pretty similar to the one we had before which are all going to be under the same workspace and you can add more and more agents here. So you can say hello. Let

agents here. So you can say hello. Let

me just put it here. You'll see the agent here. And you can add another one

agent here. And you can add another one here. Hey.

here. Hey.

And you start adding more and more agents which will work simultaneously under the same project and workspace.

Okay. I'm just going to delete the conversation right here. And delete this here as well. And this is great because if one agent is doing something and it takes a bit longer, you can start working on the next task at the same

time. Now talking about running

time. Now talking about running different agents altogether there's a concept of pricing and context window length. What that means is that yes

length. What that means is that yes anti-gravity is free but at the same time there are limits to how much power how much context you're giving the agent and how many agents you can run altogether because you can run into rate limits which means that it's telling you

hey you're using different things too much. You basically you're running to

much. You basically you're running to putting too much information to the system and you're giving too many tasks to do at the same time. You have to slow down and in that case you would have to upgrade or you have to wait 5 days.

Okay, bear that in mind when you are building this. Maybe it wasn't a thing

building this. Maybe it wasn't a thing two weeks ago, but it is now. All right,

so now on to the fun part, which is actually building the vibe coded app step by step. The first thing you want to do is you want to open folder, and you want to name a folder in your computer. You can name this uh finance

computer. You can name this uh finance dashboard. Dashboard right here, create.

dashboard. Dashboard right here, create.

And you want to press open. So this is where we'll be adding all the files that we need. Now, for every vibe coded app,

we need. Now, for every vibe coded app, you can think of it as a GPT where a GPT maybe does a specific function. In this

case, a GPT uh builds you an app. For

every GPT, you need a prompt. You need

instructions. And so when I build a vibe coded app, when I build an app, which in this case, let's build a finance dashboard, I want to give the GPT, which in this case is the agent, instructions that it needs to follow when making the

app. But I also want to give it the

app. But I also want to give it the brand guidelines of my company. So that

with these two together, it then gives us a much better output compared to what I actually want and what I'm envisioning compared to if I don't do that. Okay. So

the first thing I want to do is I want to go here and I can press new file. So

again, these are the files that the agent is going to be using when making the app. So you want to give it context

the app. So you want to give it context as to what you want it to do. [music]

And the file we name it as Gemini MD.

Now MD stands for markdown which is a type of way that you write things. So

computer can you know they can look at the hierarchy of text. Heading one will be hashtag, heading two will be two hashtags and so on. Right? That way then understands the hierarchy of text. So

that's going to be for this system instructions. And then we want to create

instructions. And then we want to create a new file which is going to be called brand guidelines.

MD and this is going to be the file where we add the brand guidelines for our brand so that it uses that when making the app. First thing we want to do here is go to the notion document which has both the system instructions prompt and the brand guidelines. If you

want access to this document, you can check it out in the second link down below in my free school community. Um,

and once you get on this document, you can simply just copy the system instructions and bring it here and paste it. And make sure that you press command

it. And make sure that you press command S, so you save the file. [music] And the same thing with brand guidelines. go

back here and you copy and then you paste here command S. So first thing is the instructions. So we say build a

the instructions. So we say build a personal finance dashboard. And by the way, these asterisk here are just bolding the text, right? So it

recognizes that that's one of the most important parts of that sentence. Uh for

tracking income, expenses, savings, and budget. This app allows users to monitor

budget. This app allows users to monitor their financial health, view transaction history, manage cards, and track spending against budget limits.

Navigation structures, we have the top navigation tabs, the header elements, the dashboard page. Uh, we have card one, card two, income, expense, total savings. It's a very detailed prompt

savings. It's a very detailed prompt explaining exactly what I want inside the software. Okay? The more explicit

the software. Okay? The more explicit you are with the instructions, the better it is. Cuz yes, these softwares are amazing in building software, but they're only as good as the instructions that you tell them to. Cuz let's say you

had to use him bolt, really fast guy, right? But if you put him in a race and

right? But if you put him in a race and you tell him to run left, he will run left and he'll lose. Why? Because of the instructions. Okay? So, it's not about

instructions. Okay? So, it's not about the powerful. And so you want to make

the powerful. And so you want to make sure that your instructions are dialed in which is why this prompt is very very detailed and that's going to be for the system instructions. And then we have

system instructions. And then we have brand guidelines. So here I put brand

brand guidelines. So here I put brand name JM solutions and then we have the logo usage, the logo. Then we have primary colors to dark blue. Uh what

kind of actual color do we want inside?

Second colors, neutral colors, typography, UI elements. So a bunch of things that I want according to my brand and my agency. Then we also have tone of voice and contact as well. Okay. So

these are two different things that again it's going to use as context. And

I want to go here. I want to make sure that this is planning. I want to make sure that this is Gemini 3 Pro cuz I care about the looks. And right here, I can say, I want you to go to the Gemini.md file that's in the project.

Gemini.md file that's in the project.

And I want you to follow like actually read the prompt thoroughly and make sure you follow the prompt step by step and build out the software that I want you to build. And also, when you do build

to build. And also, when you do build the software, I also want you to look at the brand guidelines.mmd file that's also in the project to make sure that you get the brand guidelines right for the software that you're building. Okay?

And in case you're wondering how I just spoke and it wrote this, I'm using Whisper Flow, which is a very good software that I use every single day until my credits run out. So that's it.

I'm going to press run. And now we get to see the agent live. So you can see here it says we analyze Gemini.m MD and analyze brand guidelines, which is exactly what we wanted to do, right? As

context. And so usually I make sure that I have all these things kind of lined out uh before we make the actual software. All right. So right here on

software. All right. So right here on the left, you can see that now it's making the task. So a task is basically a list of steps that it knows that it just plans that it needs to do to be able to get to the outcome which is a

full tested app right um that works which is exactly what we wanted to do and so now we'll go through step one step two step three step four and step five it's basically like a software engineer when they have to make some sort of software they will set up a plan

that they have to follow and this will be the task that he needs to do now in terms of the plan to be able to build this [music] we have the implementation plan which is just made as well and the implementation plan would be some sort

of um plan here, some sort of document that outlines the different things that are going to be within the software and the plan of action that he needs to use when making the software. Now, one

feature that I think is really interesting that they actually added is the ability for you to comment on the plan and it will use that comment when actually making the software. If I go here to let's say apply global styles

for base background, I can then comment and I can say make sure that the background is clean and it's very neutral and it's nothing crazy. I can

press add a comment and now I press review and now it will be using that exact comment when it's making the actual software. Press review. I can

actual software. Press review. I can

press submit. So if I put myself on the left hand side, you can see here that it said pending comment here. Commented on

implementation plan and so now it knows that we commented and it will start maybe iterating that feedback right that we give it uh when making the actual interface. As you can see here as well

interface. As you can see here as well while I was talking it made different files. So these are all part of the same

files. So these are all part of the same sort of software. And while I was talking, it actually started making the different files here. So these are the artifacts that I'm talking about. This

is all code, right, that we're not going to understand, but it's all going to be stored under the same project. And we

still have brand guidelines, and we still have gemini.md. And so all of these are going to be stored in that specific folder project that we mentioned that we made. And if I go back to task, I can see that now it finished phase one, it finished phase two, it

finished phase three, and it's going through one step by step. It's great

because I get to see exactly the whole thing behind what it's doing and why it's doing it. It just finished phase three. It's then going to phase four

three. It's then going to phase four feature implementation and it's walking you through exactly step by step what it's doing. And as you can see, I can

it's doing. And as you can see, I can simply just go off have dinner or have a drink and when I come back and maybe in 5 to 10 minutes, it would all be done.

But of course, there are times where it starts asking you questions as to whether this is right, whether this is wrong and you have to be there to give some input. All right, so it just

some input. All right, so it just finished the phase 4 feature implementation and I'm having fun actually looking at what it's doing and and so on. Not that I understand most of it, but you know, still nice to look at.

All right, so now it said that it finished phase 5 as well. So we had phase one until phase 5 and there's only two things that I want to show you. The

first thing is me telling it, hey, can you actually test it on the browser and make sure everything is working right?

Um, yeah, just test it and let me know if everything's all right. I can press go. And now what it's doing is it's

go. And now what it's doing is it's going into the actual local host, right?

Launching browser, which will be right here. And now it's trying to launch the

here. And now it's trying to launch the application onto our browser to make sure that it test and make sure that everything is good. You can see the blue borders on the left. You can see that it's saying get DM. Uh so it's just testing things and now it's going to

test the different functionalities of the whole dashboard to make sure that it all flows together and it does everything. Now this basically replaces

everything. Now this basically replaces the job of someone having to build a software and testing it out and see if it works and debug it and so on. Now is

it to a place where it can replace a full software engineer's role? probably

not now yet, but it will be in a place where it would just because it will pretty much do the whole process from the ideation to the implementation to the planning to the debugging to the testing. It will do everything on its

testing. It will do everything on its own. So now it says here that I've

own. So now it says here that I've tested the application in the browser and everything is working correctly.

I've included a recording of the verification session in the walkthrough which I can open. If I go down here, I can see that in the walkthrough which is one file in the artifacts, it gives us

the video of it testing, right? So you

have everything here all at the same time which is pretty nuts to see to be honest. So you get to see the whole

honest. So you get to see the whole testing process that it went through on 2x pretty much. And the last step is actually testing it on our browser to make sure that it does work. I can say can you give me the link and actually

hosted uh hosted locally. So I can press in the link and it takes me to my browser directly. I can press go and

browser directly. I can press go and here we get the link to your local hosted application. I can press the link

hosted application. I can press the link which will take me here and this will be the full vibe coded app that we just built. We have the dashboard. We have

built. We have the dashboard. We have

transactions. We have wallet and we have goals. So dashboard will have total

goals. So dashboard will have total balance, total income, expenses, savings. We can have my cards, total

savings. We can have my cards, total income. So all the dashboards, the

income. So all the dashboards, the budget, uh we have monthly spending limits, recent transactions, which are great. I can go to transactions. I can

great. I can go to transactions. I can

add a transaction like I did before. I

can press income. Let's say I do 100,000. I can do AI agency category

100,000. I can do AI agency category will be let's just do education. I can

press save transaction. If I go to dashboard, I can now see that the total balance is 104,000. So, we can see that it did work. In the wallet, we have our cards and for goals, we can set goals as well. In this case, we have vacation

well. In this case, we have vacation fund, but you can then add more and more goals. So, overall, from that one prompt

goals. So, overall, from that one prompt and those two pieces of information that we gave it, which is the brand guidelines and the gemini.md, which is the full system instructions, it was able to build this full uh working

vibcoded app that we can now host on Versel. So we can now import it into

Versel. So we can now import it into GitHub and from GitHub you actually host it on the web using Verscell so that you can actually start using it right away.

And now you may be asking yourself but what if my backed app just doesn't work or maybe there's some feature that I press it doesn't do what it wanted to do. Well in this case let's say we have

do. Well in this case let's say we have welcome back Alex. Of course my name is not Alex. I can go here to the

not Alex. I can go here to the anti-gravity and I can say hey on the first page in the dashboard it says welcome back Alex. My name is not Alex.

My name is Mik. I got the spelling wrong, but actually it doesn't. So,

there you go. Uh, I'm going to press go.

And now what it should do is it should update the actual app. In this case, if yours is a feature, so if it doesn't add the transactions, you can say, "Hey, I've tried to add the transactions. It

doesn't work. Why does it not work?" And

it will run through the whole process again. It will iterate based on your

again. It will iterate based on your feedback. Again, you're editing the

feedback. Again, you're editing the actual backcoded app to be able to fix it. Or you could as well go to the

it. Or you could as well go to the actual review, go across whatever you didn't work, so in this case, transactions. And then you can leave a

transactions. And then you can leave a comment say hey transactions doesn't work and you can add a comment here and then it will use that comment to be able to make those changes. All

right. So after waiting for the change in the dashboard to Mikuel. If I go here to the dashboard I can see that the dashboard changed just a tiny bit from the one that we had before. They added

some stuff on the left. They added the admin on the right. I think it actually looks better than before. It looks more concrete. Um and now it made sure that

concrete. Um and now it made sure that the names is Mikuel and it's not Alex even down on the cards here. And it's

great for us because it actually made it more concrete. But if in case you don't

more concrete. But if in case you don't actually like the changes that it made, you can always go back here and you can press this little arrow, which means undo changes up to this point, which basically goes back to the way that the software was before you asked for the

change. And the final feature that I

change. And the final feature that I want to show you on how to set up is the MCP. Okay, so now I'm back here on the

MCP. Okay, so now I'm back here on the default setting. I can go up here to the

default setting. I can go up here to the MCP servers. I can then go to manage

MCP servers. I can then go to manage MCP. And now I want to connect N10 to

MCP. And now I want to connect N10 to anti-gravity because if I search here N10, nothing actually comes up. So we

have to do a what we call a custom API call or a custom sort of [music] um MCP configuration. Once you get to the

configuration. Once you get to the manage MCP servers on the top here you will have this page that looks like this. I'm going to put this here and you

this. I'm going to put this here and you will want to press on the view raw config. Okay. So that you have something

config. Okay. So that you have something that looks like this which is empty.

What you want to do now is you want to go to N10. So let's go here to N10.

You want to go down to the settings and you want to go to MCP access. And you

want to make sure that you go to the access token. Make sure to refresh the

access token. Make sure to refresh the access token. And you can press here.

access token. And you can press here.

And now you will see the JSON file. Make

sure not to share the JSON file with anyone because then it will be able to use your account. I can copy this. I can

then bring it back. I can paste it here.

And now I can go to file, save. And if I go to manage MCPS, I can refresh the MCPS. And I should be able to see the

MCPS. And I should be able to see the N10 MCP connected to our account. Fast,

quick, and easy. Now, what does this actually do? Well, when you're making

actually do? Well, when you're making the vibe coded apps or vioded anything, you can now connect or you can tune in to end to receive data or to actually send data for whatever reason you wanted to in case you wanted to build an

automation around any sort of app that you're building. So, as I mentioned

you're building. So, as I mentioned before, I'm going to leave all the resources in a second link down below the actual prompts that I used to make the Vivecoded app so you can copy and paste them for yourself in my free school community. And now I hope you

school community. And now I hope you understand anti-gravity at least the basics and the fundamentals but also how to build a vibe coded app step by step which means that you're now ahead of most people who are using these kind of softwares. And if you're looking to work

softwares. And if you're looking to work with me onetoone to be able to start your AI agency then I left a link down below which is a quick video walking you through exactly uh the details of how it works and check out this video up here where I show you exactly how you can

build autonomous AI agents and workflows just by using the one and only clock code. With that being said, I hope you

code. With that being said, I hope you found value from this video and I'll see you in the next