Antigravity + Claude Code Is The Craziest Combo EVER! (It’s Insane)

The anti-gravity in cloth code combo is a complete cheat code. And in this video, I will show you exactly why that is the case and how to use it. First,

I'm going to walk you through setting up clo code inside Google's under gravity.

Then, I will show you why the setup is so powerful and how to combine it with Gemini 3 so you can get the best of both worlds. CL code for precise execution

worlds. CL code for precise execution and production grade code and Gemini for speed and planning. This way each model is used exactly where it performs the

best. So you're not wasting tokens on

best. So you're not wasting tokens on tasks that each one isn't optimized for.

And in this video I will show you how you can use both of these two tools to build a full stack calorie tracking app with user authentication and integrated stripe payments. So, if that sounds

stripe payments. So, if that sounds interesting, let's get started. Now, I

want to explain a little bit why we're going to use anti-gravity with clo and we will not just use the standard anti-gravity setup. And that is because

anti-gravity setup. And that is because when using anti-gravity and the models that it has available, think of Gemini 3 cloer.

These motors are very good for speed and large context windows. Especially Gemini

3, which has a 200,000 context window, which means the motor that undergravity has access to are very good for planning how we're going to build the app that we want to build. While if we come over

here and we take a look at cloud code, cloud code is actually very good at executions because it has very good prompts and guardrails that are excellent for advanced coding. Which

what that means is that this is basically for executing the plan that we generated with Gemini 3. So that is the setup that we're going to follow in this video and that is we're going to use

anti-gravity together with cloth code.

And now it's time that I show you how we can do that. Now the first thing we have to do is to obviously create an anti-gravity project. In order for us to

anti-gravity project. In order for us to do that, we simply have to come over here and select open folder. And then we need to come over here and select the folder in which we're going to save all the code that we generate with and

gravity and clo code. Now for this example, I'm actually going to come over here and generate a brand new folder which I'm going to call calorie tracking app and I will simply click on select

folder which opens up a brand new anti-gravity project. Now, as you can

anti-gravity project. Now, as you can see over here, we do have access to all the anti-gravity models over here. But

what we don't have access to is cloud code because cloud code it's not just the cloud 4.5 son model. It is actually all the prompts and guardrails that I have wrapped around our model which

makes it way better at going and writing code. That's why we need to go and add

code. That's why we need to go and add it in into this under gravity project.

Now the good thing with gravity is that it is basically a VS code fork which what that means is that we also have access to all the extensions of VS code and to access the extensions all we have

to do is to come over here to the left and click on the extensions tab over here which opens up this interface and as you can see we can already see cloud code but if you cannot actually see it simply come over here says for cloud and

you will be able to see the cloud code extension so simply click on it and make sure you come over here and you install it. Now I have already installed it. But

it. Now I have already installed it. But

if I actually uninstalled it just to show you how you can do it. You simply

come over here and you click on install.

Then you just wait a little bit till it is fully installed. And now you have successfully added it in in your project. So now we can actually come

project. So now we can actually come over here and close this extension and navigate off the extensions tab. But now

you might say Damian sure I added in an extension but how can I actually use cloud code? Well, in order for you to be

cloud code? Well, in order for you to be able to use cloud code inside anti-gravity, all you have to do is to come over here and turn on the terminal which is going to appear down here at the bottom of the page. And to access

cloud code, all you have to do is to simply come over here and type in cl and click on enter. And as you can see, this is going to start the cloud code extension, which we can see from over here. As you can see, it says welcome

here. As you can see, it says welcome back, Damian. Now, when you first open

back, Damian. Now, when you first open this up, you will be prompted to first set up your billing. You either select either a cloud account or you give it your claw API key and then you will see

the exact interface that I'm seeing right now. But remember the first thing

right now. But remember the first thing that we're going to do is not use clo because we first want to create a plan for our color tracking app which we're going to use by using the Gemini 3

model. So to do that I will simply come

model. So to do that I will simply come over here in the agent tab and type something like hey I want to create a color tracking app in which the user will be able to upload an image of his

food. And then this app is going to go

food. And then this app is going to go and analyze that image and tell the user how many calories are in his food and a lot of information about the macros that are inside it as well. Then I want this information to be stored in a back end

which I will tell you how you're going to create in the next prompt and based on the information that is stored in the back end. I want you to create me an

back end. I want you to create me an analytic section where it will show analytics about that person's calorie intake for four different time frames.

It should be daily, weekly, monthly and yearly analytics. Again, I will tell you

yearly analytics. Again, I will tell you how you're going to build this back end in a bit. For now, just focus on building the other functions of the app.

Make it look super modern and like a professional design team designed this app. And now that I typed this very

app. And now that I typed this very detailed prompt, what I'm going to do is I will come over here and make sure I'm in the planning mode. So, the agent can use Gemini 3 and create a plan for exactly how it is going to go and create

this app. So, I will also make sure I

this app. So, I will also make sure I tell it just create a plan. Don't build

the app now. Just go ahead and create me a very detailed plan. And now that I'm finished with my prompt, I will simply come over here and click on enter. And

as you can see, Gemini 3 has started running and right now it is going to go analyze all the information that I gave it and create a very detailed plan that we're then going to give over to cloud code so it can go ahead and execute it

for us. And the good thing with this is

for us. And the good thing with this is that we're using the Gemini 3 model which is currently completely free to use in order for it to go and analyze and eat up a lot of its tokens to generate a plan and then we'll basically

give the plan already made up to cloud code. So I can go and create the app

code. So I can go and create the app without it having to waste a lot of its valuable tokens. And as you can see it

valuable tokens. And as you can see it just finished creating the plan. So now

what I'm going to do is I will come over here and say what I want you to do is to create me one folder that includes the plan. So I can then go and implement the

plan. So I can then go and implement the plan with cloud code and then I will come over here click on enter. So as you guys saw this is the method that we're going to use to basically let Gemini 3 communicate with cloud code. We will

basically tell Gemini to go and create a folder that's going to appear somewhere over here with basically all the plants instructions. So then cloud code can go

instructions. So then cloud code can go look inside the folder take a look at the instructions and then go and execute them one by one and boom as you can see it just finished and it says that I have

created a folder named AI plan that basically includes all the instructions that it generated over here. So I will simply come over here click on accept changes and now it's time that we actually go and use cloud code in order

for it to go take a look at this plan and actually implement it. So to do that I will simply come over here and say something like hey what I want you to do is to go and look inside the AI plan

folder where you will see a plan for a calorie tracking app and I want you to go and execute it and create me this app based on that plan. And now that I typed this instruction I will simply come over

here and click on enter. And as you can see, GLO code is thinking right now. It

is searching for the AI plan which just found right now it is going to read it.

And then it is going to go and start implement all of these functions for us.

And boom. As you can see, it went over and read all those files. And now it is asking me whether I want to proceed. So

I will simply come over here and select yes. And don't ask again for npm create

yes. And don't ask again for npm create commands. So I don't want to have to

commands. So I don't want to have to manually approve each command every time. So I will simply give it access to

time. So I will simply give it access to go and execute all of them. And I will click on enter. So now cloud code is going to go run the commands in our terminal and also write all the code that it needs to in order for it to

create this app. And boom, cloud code just finished executing. And as you can see, it tells us they successfully went and written all the code for us inside this gallery tracker folder which if I

actually open, we can basically see all the code that they wrote for us. And now

just to run this app and see how this actually works. I will come over here to

actually works. I will come over here to Gemini and I will say, "Hey, give me the local host URL so I can run this app."

And I will simply come over here and click on enter. And now that we run this command, we can click on the local host URL over here. And we can take a look at our app, which as you can see guys, this

honestly does look even better than I thought it would. It honestly looks super clean. As you can see over here,

super clean. As you can see over here, we have our total calories, our protein, our carbs, our fats. some very nice graphs also showing us this information and all the meals that we ate today.

Now, obviously, this is mock fake data since we haven't added anything in yet, which we're going to fix in a bit. But

for now, let's go over in the analytic section where as you can see, it is a fully functioning section as well with weekly trends and also some charts over here. Same with the monthly and same

here. Same with the monthly and same with the yearly. And I think you guys can see why we use cloud code because basically if we had to use any other AI more than likely these sections wouldn't

actually be functioning while with using cloud code it did manage to go and create all of them in just one simple prompt. But now it's time that we

prompt. But now it's time that we actually go and we test out to see whether we're able to scan our food and so the app can tell us our calories and our macros. So I went over to Google and

our macros. So I went over to Google and search for a healthy meal. And let's say we have this healthy meal over here. So

I will simply come over here and download it. And then I will come back

download it. And then I will come back over in our app. I will click on scan food. Click on upload an image. And I

food. Click on upload an image. And I

will simply upload over that image. And

as you can see it is currently analyzing the food. And boom. Over here it gave us

the food. And boom. Over here it gave us the analysis that is 320 calories, how many grams of protein, how many grams of carbs and how many grams of fat. And we

have the option to scan another or to save it into our logs. Whereas you can see all of our logs did get updated over here. Now, obviously, we do have an

here. Now, obviously, we do have an error over here, which is definitely something that we will need to fix, but that's very easily fixable and not that big of a deal if I'm going to be honest.

But now, what I want to do is build a fully functioning back end for this app specifically. So, we can have first of

specifically. So, we can have first of all user authentication, which will mean that in order for someone to be able to use this app, they will first have to go and create an account and then they will be able to use it. and also a back end

where it stores all the information about its user. How many calories he ate every day, his proteins, his cards and his fats. So that is exactly what I'm

his fats. So that is exactly what I'm going to go ahead and do. Now there are multiple ways we can create this back end. We can create it with superbase or

end. We can create it with superbase or we can create it with Firebase. Now I

would personally advise you to create it with Firebase since it has a very easy integration with anti-gravity. And

specifically how you can do that is by coming over here clicking on additional options and then coming over here and selecting on MCP servers. And over here we can basically see a bunch of MCP servers that anti-gravity has access to.

And as you can see over here we have a Firebase MCP server which what that means is that if we come over here and we install this MCP server then we will basically be able to tell JI3 to go and

automatically create us a project in Firebase and set up the backing database for us. But as I said before, the first

for us. But as I said before, the first thing that you have to do is to make sure you download this MCP by coming over here and clicking on the install button that you should be able to see if you haven't yet installed this. Now, I

do want to point this out. In order for you to be able to install this, you first have to download node.js and also download GitHub in your PC otherwise you will not be able to install this MCP

server. But once you have installed both

server. But once you have installed both of these two, then you will be able to simply come over here and install it in literally 2 seconds. And once you have done that, all you have to do is to simply close off this tab and come over

here to the agents tab and type something like hey now I want to go and create a fully functioning backend database. First this database should

database. First this database should have authentication built into it. So

the app saves all the users that register into the app and make the sign up section a requirement. So every time somebody wants to use the app, they first have to sign up for an account or log in to their already existing account

and then they should be able to use it.

And after that I also want to create a database where all the users information is going to be stored. So how many calories did they eat for each specific day? So we can then take a look at this

day? So we can then take a look at this data and display it for the daily, weekly, monthly and yearly analytic section. So make sure you go and create

section. So make sure you go and create that using the Firebase MCP tool that you have access to. And now all I have to do is to simply come over here and click on enter. So, Gemini 3 is going to

go use the Firebase MCP and create this backend database for us. And as you can see, it has already started and it's going and looking whether it has access to this MCB tool, which it does. And now

it is creating the implementation plan for exactly how it's going to go ahead and create this back end. And boom. As

you can see, Gemini almost finished setting up everything. It just tells us that we need to execute these commands.

We can either tell it to go and execute in itself or we can just execute them by ourselves which I'm actually going to do right here simply because I can. So I'll

simply execute them in the terminal just you guys can also see this backend database being activated as we speak and boom. As you can see the deploy was

boom. As you can see the deploy was completed. So let's run npm rundev. Now

completed. So let's run npm rundev. Now

you don't actually need to understand what all of this is and neither do I honestly. Geminina just told me to

honestly. Geminina just told me to execute them and I went over and executed them. And right now over here

executed them. And right now over here we have the local link URL which if we actually click on and as you can see when I open it up I'm first prompted to sign in with my account or to sign up

with a brand new account which I'm actually going to go ahead and do right now. I will simply input my email

now. I will simply input my email address. I will input a password over

address. I will input a password over here and I will also confirm my password. And if I come over here and

password. And if I come over here and click on sign up, it says that it failed to create the account and that is to be expected. We also need to go inside

expected. We also need to go inside Firebase Studio and enable the email/ password login section. So if I first come back over to gravity and I say hey when I try to create an account it says

that it failed and that is because I haven't gone inside my Firebase console and enable that login parameter. So give

me the link so I can go and enable it.

Now if I'm going to be honest usually Gemini does tell you to do it by itself and you don't actually have to go and tell to give you the link. But this

time, since this was a kind of more complicated back end, it didn't actually tell me that I need to go ahead and do that. That's why I just simply told it

that. That's why I just simply told it to go and give me the link. And as you can see over here, it says, "Oh, you're right. Firebase requires you to

right. Firebase requires you to explicitly enable the sign-in methods.

Here's the direct link to your project."

So, I will simply click on this link, which is going to open up the Firebase website. And as you can see, the color

website. And as you can see, the color tracker project has already been created. Now what we have to do is

created. Now what we have to do is simply come over here and click on get started. And then we need to come over

started. And then we need to come over here in the signin methods and enable the email/ password option since that is the method that we want our users to be able to sign up with and simply click on save. Now if you want to enable another

save. Now if you want to enable another one you simply come over here you click on add a new provider. You can enable Google for example by coming over here also adding in a support email and then coming over here and clicking on save.

And now the people in your app will also be able to sign in with Google. And the

other thing that you need to do is to come over here in the build section and navigate over here in the real time database where you simply need to come over here and click on create database.

Then select the location that is closest to the location that your app is going to be at. Let's say for me it's going to be Europe. So I will simply select

be Europe. So I will simply select Belgium. And then you also have the

Belgium. And then you also have the option to start in locked mode or start desk mode. I will actually start in

desk mode. I will actually start in locked mode and I will simply click on enable. And boom. We have also created a

enable. And boom. We have also created a real-time database which means that if I right now come back over to anti-gravity and I say something like hey I went over and enable the login parameters what

should I do now and I click on enter it is going to tell me to basically restart the app. I run it again and you will see

the app. I run it again and you will see that I will be able to actually create an account and that information is going to be stored over to Firebase. As it

says over here it basically tells us to retry the signup. As you can see, I just did and successfully managed to sign up in our app. Where as you can see, all of our stats over here have been resetted to zero since we actually built the

backend database. Which means that if we

backend database. Which means that if we actually come over here and we click on choose image and we upload the same image of our food and we save it into our logs, it is going to update all of

these different sections over here along with the graphs. And if we navigate over in the analytic section, we will be able to see everything that we consumed for the week, for the month, and for the

year, which is exactly what we wanted.

And just to prove that this data is stored over in Firebase, if I come back over to Firebase and I navigate over in the user section, you can see the exact account that I just created when I

created it when I signed in and also the unique user ID of that account, which means that that user is successfully stored in our backend database that we have built with Google's Firebase. And

now the next feature that I want to show you how to build is how we can integrate Stry payments inside this app. So every

time a user signs up for an account, they first have to pay a monthly fee and then they can access the app. So to do that, we have to go over to strip.com and sign up for a completely free account. And once you sign up, you will

account. And once you sign up, you will be brought over in this interface over here. Now all you have to do is to

here. Now all you have to do is to navigate over here in the developer section and then make sure you come over here and you click on API keys. And once

you're in this section, you will be able to see your publishable API key. So make

sure you simply click on top of it in order for you to copy it in your clipboard. And then you need to come

clipboard. And then you need to come back over to Firebase Studio and say something like, "Hey, I want to integrate Stripe payments inside this app." So every time a user signs up to

app." So every time a user signs up to it, they first have to pay a monthly price of $10 and then they should be able to use the app. This is my Stripe publishable API key and then simply

paste in that API key. And then you have to say this is my secret API key. Then

come back over to Stripe. Make sure you copy your secret API key which you will be able to find from over here. And if

you don't have a secret API key, simply click on create secret key. Then click

on build your own integration and click on create secret key. Then verify

yourself, give it a name. And then you will be able to copy it from over here.

So just come over here, paste in your stripe secret API key. And now you can either tell Gemini to go and implement the changes by itself or we can also use cloud code for these changes. Now for

this example, I will actually choose to use cloud code since in this video I'll show you how to basically combine the two. So I will say create a folder that

two. So I will say create a folder that is called stripe with detailed instructions on exactly how an LM motor can go and make this feature happen.

Just create the folder with instructions. Don't go and make any

instructions. Don't go and make any changes to the code. Now I will simply come over here and click on enter. And

while Gemini is creating this folder with detail instructions for exactly how code can go and integrate drive into our app, let's come over here in a terminal and actually call cloud code. So to do

that I will simply come over here and I'm going to type code and I will click on enter. Now you can either tell it to

on enter. Now you can either tell it to go and create a folder so we can call code in order for it to go and implement stripe into our app or we can simply tell Gemini 3 to go and implement it.

Now I would actually advise you to tell Gemini 3 since this is a pretty simple thing to do. Just tell Gemini 3 to go and implement stripe. You don't really need the power of cloud code for this feature. So I will simply come over here

feature. So I will simply come over here and I will say so go ahead and integrate dry payments into this app and I will simply come over here click on enter and I think you guys can see that we don't

constantly use cloud code we only use it when we really need it which is we were first building the front interface of our app and the main functions of it in very small little changes like this like

building a backend with an MCP server or integrating Stripe payments we don't actually have to do that and just so you guys know there's an even easier way to integrate Stripe payments if you

actually want which is coming over here, selecting again MCP service and simply coming over here searching for stripe because as you can see over here under gravity also has a stripe MCP server

where all you have to do is simply come over here and paste your stripe secret key and click on save and then you can basically tell Gemini 3 to go and integrate stripe image into your app

using the stripe MCP server but for some reason I have found this to give me errors sometimes as it did right now that's why I choose to go with a manual way more often than not but there is

also an option it just in this beginning that's why it gives you some error sometimes and now as you can see Gemini has created a plan for us so all I have to do is simply come over here and

accept all the changes and I will simply come over here and say sure the plan looks good go ahead and implement it so now all we have to do is to sit back wait till Gemini goes and integrates

drive payments into our app and boom under gravity just finished integrating strip payments into our app. Basically

just pressed on accept every time that it asked me to run a commanded terminal.

And now it's time we actually go and take a look at the app. So if I actually open it up, I see this unlock premium section where I have to come over here.

I click on subscribe now. And then I'm prompted to come over here and pay with my credit card information. So then I'm able to access the full app. And now the last thing that I want to show you how

to do is how you can deploy this app over to the cloud. And how we're going to be able to do that is by again using the Firebase MCP. Specifically, we're

going to come over here and say, "Hey, now I want to go ahead and deploy this app over to the cloud." So, make sure you use the Firebase MCP with Firebase hosting and you go ahead and you deploy this app over to the cloud. And now I

will simply come over here and click on enter. And now Gemini 3 is going to go

enter. And now Gemini 3 is going to go use the Firebase MCP tool that we have access to and deploy this app over to the cloud for us. And boom, it just finished deploying the app. which means

that we can finally access it from a live website URL after we obviously pay the $10 monthly price. So that's how you can combine the power of anti-gravity

with the power of cloud code and build beautiful looking full stack apps just by typing a few simple prompts. Now, if

you want to learn how you can use this AI code builders along with also learning how to use any 10 to build amazing apps and amazing workflows that automate real business processes and

also how you can sell these workflows to businesses and start making real money with your AI skills. Click the top link in my description and come and check out my school community in which I show you

how to do exactly that. And

specifically, I have a 100 day road map in which I show you how you can get your first highpaying customer by selling workflows than the one that I showed you in this video. And if you join the community, you don't manage to land the

first high-paying customer, I just simply give you all of your money back guaranteed. Since I'm certain that if

guaranteed. Since I'm certain that if you follow this 100 day road map, it's almost impossible not to land the first high paying client, which basically means that you either join the community

and you at least 10x your investment or get all of your money back. guaranteed,

which is a pretty awesome offer if I have to say. Plus, you will also get access in my AI SAS road map course in which I show you how you can build and grow your own AI SAS and also you get

access to the build a $1 million SAS series in which I document the process of me building my own AI SAS exclusively inside the community. So, if you want to get access to all of this content along

with getting access to a community of more than 140 members that are interested in AI and most importantly making money with it, make sure you click the top link in my description and you come and check out my school

community and I will be waiting to personally help you immediately when you join. But also, YouTube thinks you're

join. But also, YouTube thinks you're going to enjoy this video in which I build a full stack app using just anti-gravity in Gemini 3. So, make sure you go and watch that next and I will

see you over