How creators use Zapier to build no-code AI trading bots

And there we go. We have successfully set up an AI agent that will automatically trade on our behalf. We could be sleeping, we could be cooking food, doesn't matter. It's trading looking pretty good.

Let's build an AI agent that's going to be able to trade stocks on our behalf. Your next question might be is, Corbin, why is there a bunch of politicians behind you? Is because this AI agent we're going to train on US politicians trading data and trade off that. The best part, though, is that this is going to require no code, all automatic, and let AI do the heavy lifting of the actual trade. And on top of that, you're going to be able to laser in this AI agent to trade on whatever you care about. So if you're like, Corbin, I don't really care how US Politicians trade, then that's fine. You're going to be able to see how you can laser this in for whatever you think is relevant for a trade.

Sound good? Let's jump in.

Welcome back. Yo. In today's video, we're going to be creating a Algo trading bot that trades on our behalf. And the reason I have that little alpaca behind me is because this is what's going to give the AI agent the ability to trade stocks. So instead of you sitting at your computer like Corbin, I have to hit buy and sell. No, no, no. We're going to let the robots do that.

And the best part is I'm going to show you how to do this with no code, all automatic, using Zapier agents. I've done tons of videos on Zapier agents, so you can check them out. I'll leave a couple in the description that shows you how to do web parsing. I believe another one there will show you how to automatically respond to emails. But one thing as I was playing around with this, that this can do, which is awesome, email is basically be a trading bot.

So before we dive any deeper, this video is sponsored by Zapier. And this is part of my ongoing series with Zapier, where we keep jumping in calls and we're like, yo, what's really cool? What do you want to do? And I'm like, you know what? Let's show them how to create a trading bot that could be used on Wall Street.

Sound good? Let's jump in.

Last I know this video is kind of inspired by Nancy Pelosi stock trading because, like, there's like a whole community behind this. You might not know this, but there is like a ton of websites and community around just Nancy Pelosi as a politician, not all the politicians, which I'll show you how to do in this video. But in theory, if you just want to track Nancy Pelosi stock trades, you can with the method I'm about to show you in this video.

Enough talking, let's do it.

First thing we need to do is sign up with an account on Alpaca. I'll leave this in the description down below. Sign up for free.

So here we go. Once we are logged in, this is going to be a paper account. Obviously to trade real money, you'll just essentially create a real account, connect bank account, etc. Bro.

Now though, all we care about is seeing how we set this up. So. So let's get our API key. It does look like we'll need to enable an authenticator, so we'll do that. Activate. If you don't have an authenticator app, just go to your app store, type in authenticator app, you'll be able to download one for free.

So now that we're logged in here, we're going to generate a new key for the paper account. Obviously you would generate the new key for the live account when you're ready to go. I got my key right here, so I'm going to go to copy. I don't know if I'm going to blur that out in editing. I might not. But if it's right there, you can go ahead and try to use it. It probably won't work.

So once we have that though, we're going to come back to Zapier. Go ahead and go to the app section on our sidebar here. We're going to look for Alpaca. It's the first result, but we can just go ahead and search it anyways. Alpa, go in here. I'm going to delete this old one because this is like as you can see a year ago. So we're going to do a new connection here to add connection. And there we go. It's actually really nice here. So we can go ahead and just hit paper. Allow. We are connected.

This part right here is fundamental as this is the part that's going to allow us to actually trade the stock or sell the stock. What's really cool about a packet too, you can do this with crypto. You can even do this with option contracts as well, if you like. That want to be a little risky? Sometimes you gotta be risky. I do like options.

Now that we set that up, let's set up our AI agent. Yes, our moneymaker. Our really Good time maker. Let's go ahead and call this agent Algo trader. Start from scratch.

One thing I love about zapier agents that is really hard to access throughout the entire market is its ability to web scrape any page. So traditionally, if we were to do the logic I'm about to show you right now, we would have needed like an RSS feed in order to identify when new data is available. But what you'll notice is typically RSS feed feeds aren't really that readily available. Now I did a whole other video showing how to create a trading bot that's based off stock news, just like general RSS feeds. But this one we're going to make a little bit fun and we're going to trade based off US Representatives.

So here is what we're going to do. We're going to go to do run behavior here, create behavior. Let's start off untitled behavior. We're going to say this U.S. rep. Trade, call it that nice instead of on demand. Because we want this to work. Whether we're asleep, away from our computer, or maybe just in the Bahamas on a beach just drinking a pina colada, we're going to do scheduled. And the way we can do it is we can do every day, time of day we'll do on market open. So for me it's around 8:30am for you would be whatever your time zone is. Also this is US Markets now because the markets aren't open on the weekends, we don't care. So we're going to trigger on weekends. No. Nice.

So here's the situation as I drink this amazing Houston blend of coffee. Nice little cinnamon in it. So we're going to build this step by step. And while we're building this, we're going to be testing it.

Here is step one. Here is a website link of the recent trades made by US representatives. What I want you to do is find the most recent trade and output the stock ticker and US Representative website link and then provide the relevant data here. The way I got that website link is I simply just went to this page and as you can see, this shows most recent trades. And then I come up here and copy the underlying link here. Paste over. This method can be applied to any website, any context, anything you care about, follow the same step.

So let's first off see if this even works. We should be expecting an output here of French Hill in cvs. Start instructions and test.

I want to point out as well, in reality, the only thing we really care about is the stock ticker. To make the trade, I also included the US Representative's name because that gives you more context of maybe things you care about. Maybe there is certain US Representatives that you don't care about their trade. So just ignore that and you'll see how we can ignore certain things depending on the flow past this.

There you go. Successfully passed stage one here, where I identified French Hill and CVS as the most recent trade. And obviously with the stock market and how the stock market works, timing is very important. That can make the difference between a trade you make a hundred percent on and a trade you make like 12% on. That's why I like options.

Okay, let's do this. Because I don't really care about the U.S. representative. If you do put it there, I'll just do stock ticker. And then next step would be this. What I'm about to show you right here is going to work. And when I did it, I was like, this is awesome. Like, you can apply this to anything based on the result, check if it is a previously traded stock here.

How do we do that? Corbin, watch this. We're going to use an action called storage. And the first thing we need to do is we need to see if this was already previously set. Right? To do that, we are going to do get value.

What is storage? How do we leverage this? I want you to think of storage as this is like our little notebook of like, okay, I already traded CVS two weeks ago. Therefore I don't necessarily want to do another trade with CVS for XYZ reason. This kind of logic can be applied elsewhere. But the idea is this. Set a specific value. We're going to store data towards a key, right? So we're going to say US Politics, trade. And what happens is that we'll create an array here. When I say array, this is like data point 1, data point 2, data point 3. The first data point that will be put here will just be the ticker. This will make more sense as we get going here. But as you can see, if no search results are found, we're going to do this. Set specific value for this field and we're just going to mark as successful because if no search results are found, that means that we haven't traded this stock before, which means that we're greenlighted for the next stage.

Here, watch this. So that's getting the value, we're checking it. But that means that if we haven't traded this stock before and we haven't stored it in that key before, let's go in and set it then. So we're going to do another action here. We'll do storage again. This storage block is like extremely underplayed with zapier. Like a lot of people don't even realize it exists. Trust me, you can really create some cool automations with this. I'm going to have to do a whole separate video on it. But for now we're going to do get value because we're not setting the value or sorry, we're going to set the value because we already did get value in the previous step here. We're going to set value here.

And one thing I want to point out is I'm going to say based on the result, check the previously stocks traded here. If it's there, then exit out of all actions because we just want to leave the trade. Like we don't want to actually trade it. We've already traded it before. Let's just leave the situation. If it doesn't exist in this storage. Let's go ahead and set this ticker here. And this is going to be setting a value. What's very important here is that we're going to be setting it towards the same key US politics trade. This has to be spelled exactly like that. So we're going to say set specific value. Boom. And then we can just let the agent handle the actual value he's going to put in the field because that's going to be a variable depending on the ticker that is found within this flow save.

So this makes more sense here. I'm going to add one extra line here just so we can see clarity of all this. Like this, all this working together. Right here's the situation and here's what we're going to expect right here. If this is a new ticker that we set, so say yes, new ticker in the ticker name. If it's an old ticker that we've already set, say we traded it already. Therefore what we should see in this first runaround is going to be yes, new ticker retest behavior because we've never traded and stored that value in cvs. But on this second time around it's going to say we've already traded it already. This kind of logic that you're about to see right now, this can apply to a lot of stuff.

We should also see the thought process here from our AI agent basically being like, yeah, look it, let me check if we've seen this ticker before. Like is this not awesome? This is cool. And then obviously it hasn't Seen it yet. So the next step is like, okay, we haven't seen it. Let's store the ticker. Then storing it now in testing, it's going to approve whether or not this is a correct way of handling the information, which obviously is because that's the ticker cvs. And then we should be saying here, yes, new ticker and the ticker name. Talk about like, next level. People are underplaying agents right now. There is so much you could do and that's just cool. Let's keep going.

Actually, before we get going here, let me show you how advanced these agents are and really gives you context of what you can do here. Retest behavior. CVS already exists in that data storage. Now, therefore, this is going to say, we traded already. Oh, yeah, oh, yeah. We in the future. Now we traded already. The most recent trade was cvs, but this ticker was already in our storage. I like it.

So knowing that we may have to just change our prompt here from like the most recent trade to. To like the second most recent, actually, what we'll do then is we'll just train your storage key. So, like, let's say you're building a bunch of tickers and after a while you're like, you know what, Corbin? It keeps like overlapping and I'm having issues. Just add like a one there. It's gonna be a whole new storage bucket here and it's gonna be able to reference new trades. Think of it like cleaning the slate.

Now, here's what's cool. Now that we've done that, let's do some research. This logic right here is malleable. Approach this how you want to approach it for your threshold of either buying or selling, slash shorting. And obviously the other stuff, when it comes to option contracts and everything of that nature, it's simply this, though, based on this stock ticker, I want you to do more research whether the stock is a buy or sell. From there, you can add more context of like, only buy it if the floats this much, only buy it if it's up 10%, etc. Based on your research, give me a buy score out of 100. If it's over 70, then we'll say it's a buy and do. And then I'm placing the action here from earlier from Abaca Place order. So we're going to use your apocalypse account here for the symbol. We're going to let the agent generate that because it knows a CVS or whatever the ticker will be for the side. I went ahead and just Set the specific field to buy. So it just knows to buy quantity, you're definitely going to have to set this. I just put 10. Are you trading a hundred? Are you trading a thousand? Your discretion. And that's like stock amount, how many stocks you're buying. Notional agent. And then the type could be limit depending on your context. I'm going to just say market. Everything else is the agent, hit save. And then if it's below 70, that do nothing. Let's just exit out all actions because this is not signaling a buy based off the AI agents research.

Watch this. No coding, just prompting retest behavior. Algo trader is on it. Loading. In the website that we provided as data, capitaltrades.com trades. Checking if it's already been traded before in the past. And because we cleared that storage, it's not going to find it. So now it's going to store it nice and it's going to start analyzing if it's a good investment. This little approve situation won't pop up when we actually turn it on. This is purely just because we're testing within this little environment. So we're going to say approve for now.

Here we go. Now it's going to be researching whether this would be a good buy. And based off our scoring, it should be over 70 in order to incur a buy. And that score can be different. Right? So maybe your threshold is 50, maybe it's 95. Whatever it may be, set that threshold. Ooh, okay. It received a 58 out of 100. So this was not a buy. I guess you saw the version of it not placing that order.

Let's make it so it actually does place the order, though, just so we can get an idea. So I'm going to go ahead and real quickly reset my storage buckets here so we can kind of go through the entire flow again. And because it gave it a 58 there, we're going to lower like our threshold to like, if you say it's over 40, then buy and do. I would not encourage that. I definitely want to encourage that retest behavior while that's running and loading here, you can come to your alpaca account, go to home, and this is where you're going to be able to see all your trades.

And there we go. We have successfully set up an AI agent that will automatically trade on our behalf. We could be sleeping, we could be cooking food. Doesn't matter. It's trading looking pretty good.

And one thing I want to point out that's going to be really cool that we're going to be able to add to this Algo trader in the future is our ability to actually watch the news or any videos on the topic of stocks, crypto and the markets. Pretty soon here, the platform of Bump ups is going to have API documentation available. That's going to allow us to watch any YouTube video, any newsfeed video, any type of video, and we're gonna be able to analyze it and see if the trade is good or not. API and Bump ups pretty soon.

That concludes today's video. Make sure you leave a Like if you found value in the description down below. I'll leave my other videos on AI agents that you may be interested in. Without further ado, I'll see you in the next video. Algo Trading Bot two Random Videos that's my face. I'll see you in the next video.

But for now, you know the steps and softwares required to set up a fully functional AI agent trading bot. Choose your crypto. Buy, sell. Bullish, bearish. Let's find out. Let's create an AI trading crypto bot that's going to allow us to trade anything you see to the left here, all that good old crypto anything, Corbin, anything. We're going to be able to trade Bitcoin, Ethereum, and don't worry, we're going to get some meme coins in there like Dogecoin.

If that all sounds good, let's go ahead and jump in today's video.

Welcome back, y'all. In today's video, we're going to be setting up an AI crypto trading bot that's going to be able to trade on our behalf whether we are asleep in the ocean, in the mountains, wherever it may be.

Today's video is sponsored by Zapier. As we've teamed up and we've decided we're going to hit head on a bunch of cool topics that you actually want to see. And this one right here, based off my other stock trading video, which I'll leave in the description down below, was highly requested. The amount of comments I got saying, Corbett, I need to see how to do this with Bitcoin, Ethereum, and trust me, there was a lot of meme coins in there as well. So let's go and figure out how to do this.

Best part is this gonna require no coding, just prompting. And the first major software we'll need is Zapier Agents, which is everything behind me. So I'm going to leave the first link in the description giving you the ability to do this for free. Click it, get going. The next software we're going to need to use is Alpaca. The reason we're going to use this is this is what's going to give our AI agent the ability to even trade on real markets. Therefore, go ahead and create your Alpaca free account as well. I'll leave it in the description down below.

I'm also planning on doing a video on options, so make sure to subscribe here as that will come out pretty soon. As you can see from the stock video, we went ahead, did the analysis with the AI trading bot and bought some CVs. And not to brag, but we're already up 22%. Oh yeah, that's $117 of profit. No red, only green. Let's go and begin.

Start from scratch. In our previous video, we created a stock trading bot based off how politicians are Investing in the market. In this video, we're going to create a more lasered in version of this bot, specifically giving it a edge in the sense of whether it's bullish or bearish on the way it trades. So let's go and begin here with run behavior, create. Create behavior as we need to create a behavior that's going to allow our trading bot here to actually sell and buy crypto.

In this video I'm going to show you how to do with bitcoin, but in theory you have access to a ton of other cryptocurrencies. I'll go ahead and leave a help article in the description that shows everything Alpaca has access to and that's going to be your limitations of what kind of coins you can trade. Let's do it though.

Now, when creating trading bots, there is a whole plethora of different ways of approaching the logic of what constitutes a buy in a cell. In reality, typically you're going to want to give the trading bot a bullish or bear bearish kind of edge for its underlying opinion. Therefore, the behavior we're going to give this trading bot is going to be very bullish. This is your discretion though. If you feel like cryptocurrency bitcoin, you're bearish on it for the next six months, then you would just take my same logic and reverse it. But assuming we're going to be bullish here, let's do bullish crypto.

The on demand is we're going to set a schedule by Zapier. This allows it so that me and you don't have to keep logging in and telling the trading bot what to do. No, no, just, just do something, okay? I just want you to do something. Now, the one thing about crypto that's different than stocks is that it's traded 24 7, 365 around the clock. Therefore we can do every day here and then the discretion of time of day is completely up to you. So I'm going to just do 8am this is going to run the behavior of all the logic you're about to proceed here at 8am every day. Make sure to show more options here and say trigger on weekends because we can trade on weekends, safe.

Therefore, let's set up the very first step here of what our crypto bot should do when approaching a trade. Check this website, semicolon, parentheses. I'm going to personally give the coin market cap as the thing it's going to scrape data for when it comes to pricing. In reality, this could be any website. You could quite literally put the link to someone's X profile here if you want to follow their trades. Or alternatively do what we did in the stock video where we found a very specific type of trade which was a politician trade. For now though, we'll go with the general idea of just what's the price, what's the percent, et cetera.

Now when creating these trading bots, what I like to do is every step, let's just gut check it and make sure it's actually working. So coming back over to CoinMarketCap, we should see a price around 96,418 outputted, therefore save instructions and test. Let's see if it can do it.

One very, very powerful thing about our AI agents with Zapier here is its ability to scrape data from websites. This kind of functionality is extremely valuable and really easy to access through Zapier. Okay, so it's going to the CoinMarketCap site. Let's see if we see something around 96,418. And there we go, 96,366. Because as you know of crypto, it's very volatile. If you didn't know that then I guess you just learned something new. Today we are not trading bonds and CDs right now we are trading crypto.

For us to create a trading bot that has context of its previous trades, we're going to use a tool within Zapier called storage. So we're going to say insert action and we'll do storage here. Storage by Zapier. And what we're going to choose is set value. We're going to go ahead and click this and let's go ahead and change some stuff here. First major thing we need to change is we need to set a fixed text for our key here. What do I mean? Think of this as this is where we're going to be making multiple entries of the Bitcoin price that we're going to use in later on. Logic. Another way of thinking of this is just like a very easy Google sheet. Like hey, today's price was xyz, tomorrow's price is XYZ going to be able to cross reference this. So I'm going to call it Bitcoin price scrolling down here.

We're going to want the agent to generate the value for this, right? Because obviously the price changes depending on the days we're going to hit save here. Therefore this leads to the next step here. Which means that if we have an actively live trading bot here, we need to make sure that we're always selling when we're in the positive. You know the old saying, buy low, sell high. This situation. So we need to check our most recent purchases. Same situation here. Insert action, data, table or storage. And what we're going to do here is we're actually going to get a value. Now click this and this is going to be a different key. We're going to set this key to purchase Bitcoin. This is the amount we've purchased in the past at the very specific price point. Let AI do that field and we'll move to step four.

Now, in step four, this is where we're going to have to add conditional logic. So I'm going to read it in text form and then I drew it out on a whiteboard so it makes a little bit more sense. But the idea is this. If the price you set in step two, the current active price, is lower than the price we find in step three, which is our previous purchases, then we're going to go to step five, and step five is going to be buying, because we are in a bullish edge here. This is the way we're trading this bot. But if the price in step two is higher than the price you find in step three, therefore, let's say you bought at 85, but now the price is 90, we're going to go ahead and skip to step six here, which will be a sell. And finally, we're just going to add one little caveat here. Assuming we're just starting this crypto bot, where if no price exists in our little price catalog here, then we are going to go ahead and skip to step five and just default to a buy.

Let me just walk through this logic on a whiteboard, so this makes more sense. So we have our current active price. Let's just say, say right now, bitcoin's trading for 90,000. We just got this data. It's fresh. This is going to go to step three, which is going to check our price catalog of our previous purchase price. Assuming we bought at $95,000, we're going to be at a deficit for 90,000. And because we're making this trading bot bullish, we're going to go to go to step five here, which will be buying more Bitcoin. Basically, we're doubling down because in theory, if you are in a bullish bias, you are assuming that you're going to see higher and higher prices.

Now, what's the flip side of this? If the live data reads 90, we'll go to step three. Check our previous purchase catalog and if it's 85,000, which means we bought at 85K, therefore we're in a positive trade here. We're going to go to step six here, which will be a sell. Therefore, when it comes to schedule by Zapier or Trigger here, you may want to opt for maybe a longer duration, like check every week rather than every day. But this is completely up to your discretion.

Let's get going here. Let's give our bot the ability to even do these actions. So we're going to say step five here, which is a buy, go ahead and purchase more Bitcoin using this action. And we're going to say insert action. This is where alpaca comes into play. Alpaca. And we are going to buy an order place order, let's go and click it and let's check it out. So our simple is going to be very specific field, etc, and then the side we're taking because we're bullish here is going to be set a specific field and it'll be buy. And then the quantity is obviously up to your discretion. So what's your budget, how much do you want to put in a trade, how much you want to take out of a trade, etc. So for now we'll put 0.1 Bitcoin. Everything else we can kind of leave as is and hit save. So this is going to buy Bitcoin in a trade.

So once we actually buy the bitcoin, we need to set this value in a purchase catalog. So we're going to go ahead and hit insert action. We're going to go to Storage by Zapier again and we're going to choose set value, click it and not bitcoin price. We're going to be setting our purchase Bitcoin, which is our key that we set earlier in step three. So I'm going to hit save here. And just to confirm. There we go. Purchase Bitcoin. These have to be the exact same dictation letters, et cetera, for it to reference the same data source.

So the logic that's incurring there is that anytime that we're in a deficit from original purchase price, we're going to just keep doubling down, double down, double down, double down. That's why this entire trading bot has a bullish bias. If you want to see a more like analytical trading bot, I encourage you to check out the stock trading video that I have in the description, which bases its trade not necessarily on bullish or bearish, but on sentiment in the market analysis of articles trades that are being made by government officials. Although if we're winning in the trade, we can go in and sell.

So insert action. We're going to do a packet again, come down here, we're going to say placeholder, click it. We got Bitcoin here, Nice. Instead of buy, it's going to be sell. And I would suggest using the same increments, so there's no issue there. So we'll do 0.1 save. So effectively what we just did right here is we created conditional logic for our bot to follow to a T. And that's what makes AI agent super cool. As in reality, to code this out, it would require a lot of Python.

So let's go and try it out. Save instructions, retest result. So first thing is going to do is going to grab the live active price of Bitcoin. And because of how our logic is set up here, this is going to initiate a buy because nothing exists in the purchase catalog as of now, so, so determine next step here. Perfect. So right now it's storing the value of the current price and there we go. It is currently placing the order, it's going to give me a confirmation, I'm going to hit approve. And as of now it would have executed.

But one thing to note about a packet here is in order for us to make live trades, we can't actually use a paper trading account. Essentially, when trading with Zapier agents and Alpocket API, you got to do with a live money account. So just put in a couple of dollars in there just to play around with it. Although what you'll see here is pretty cool. Our logic got fleshed out pretty nice here in the sense of that it found the current price, it stored it in Z year, it successfully cross referenced and checked whether the price was good or not. And then obviously based off our logic here, it wanted to execute a buy order. But right now, Alpaca doesn't allow buy orders or sell orders or any type of orders in a paper trading account. So to truly flesh this out and try it out, we would need to open a live trading account, put in some money and proceed.

Now, when it comes to this logic right here, one thing to note is that this could be reversed, right? So if we're bearish, we could short Bitcoin, sell on it, and do whatever you want. One, you should probably whiteboard it out on the way you want your trading bot to function and then translate that into text that you can input into a Zapier agent. But two, these branches can get very, very much more complex. In theory, we could add a threshold here where even if it is going to want to do a buy, we could add some type of blocker here that's like we'll look up current news about Bitcoin and see why the price is down or see why the price is up before making your decision. This extra layer of insurance can basically help you out in situations where maybe you don't want to trade purely on price, but add a little bit extra layer of complexity to your agent.

So for what I'm about to say, obviously I'm not a financial advisor, so this is not investment advice when setting up this bot. I'd probably opt for a longer duration rather than day to day. Unless you're dealing with a very volatile crypto. Typically you're looking at a week here. So like every Monday run this kind of trade.

But for now you know the steps and software is required to set up a fully functional agent trading bot. Choose your crypto buy. So bullish, bearish. Let's find out if you felt like you learned something. Make sure to leave a like, it's completely free and I'll see you in the next video. AI Crypto trading bot. Two random videos. That's my face. I'll see you in the next video.

Hey, I'm Mike and today I'm going to build an AI trading bot in 10 minutes without writing a single line of code. Now, if my AI bot loses money, I'll embarrass myself and get ready for this spicy pepper challenge. If the bot profits, then one or more lucky subscribers will win a free year of Zapier, the tool I'm using to make this AI trading agent. So make sure to like and subscribe to this video because it could be you winning. Remember, this is not financial advice or trading advice. I'm a noob to this.

Here's the plan. I'm going to use Zap Zapier to automate and listen for real time trading signals from Taapi IO, which provides technical analysis on not only the stock market, but also cryptocurrency. I'll then pass that information onto ChatGPT to analyze and make a decision. And then we'll use Alpaca to place a real time trade. Now don't worry if you've never used Zapier before or APIs, I'll walk you through each step. It'll be nice and easy. If you can use a web browser, then you can definitely do this.

By the way, thank you Zapier for sponsoring this video and enabling me to turn Trading Chaos into an organized workflow. Now, while you could do this for free, I'm gonna upgrade to professional. It gives me access to webhooks so I can use that technical analysis API and also allows me to run more workflows which will be helpful when I'm trading in real time.

Okay, with that done, I'm logged in. Let's create our first Zap.

Before I add my first trigger in Zapier, I'm gonna get my free API key from TA and I may as well go for this seven day free trial. You could pick your own indicators, but I think the relative strength index is a popular one and a good one to start with.

Now bear in mind here, I know nothing about the stock market. I'm trading as a complete newbie, possibly like you, and I'm going to rely on AI to interpret the signal to choose what to do. With that said, let's go to my account and generate a new API key.

Now, don't worry, you don't need to get deep into the code. You just need to copy and paste this into Zapier to get started. You just copy and paste this into Zapier.

So let's do that. And for the trigger, we're going to go for a Web hook here. You'll also notice it's a feature available to Pro accounts. For the trigger event, I'm going to go for retrieve poll and this will execute every two minutes when I'm upgraded to a Pro account. If you're on a free account, It'd be every 15 minutes. So you really can trade with signals in real time. Here. Let's click continue. We'll paste in that URL we copied from Taapi IO. Now for my secret, we put in our API key. Now for my secret, we'll add in the API key that I just generated and then we've changed it to type stocks and we're looking at the symbol for Tesla. We're going to be trading Tesla here in this example. And the interval is every one minute. So we get real time information as our zap runs every two minutes. With that done, all we need to do is enter a deduplication key, which in this case will be value. That's the value of what we get back when we make our first call in this zap. And then we click continue.

And finally we just need to test the trigger to make sure we we get some data. And there we go. We've got an entry here, entry A. The record has been pulled and we can have a look at it. And Therehere we go. ID is exactly what the value is. And this is the RSI Right now that is exactly what we're going to Pass on to ChatGPT to make a decision.

Okay. With this working successfully, we will continue to our next record, which is of course going to be ChatGPT. There it is, ChatGPT. We'll add it in. Now we need to make sure we're signed in to connect ChatGPT to Zapier. And when you do this, it gives you instructions on how to create an API key. Again, really simple.

Now we'll work with the model GPT4O mini. Now I'm going to write in the user message the current relative strength index for TSLA brackets, Tesla stocks is. And then all I need to do is enter a forward slash and then pull in the value here that we had before. So that'll be there included in my chat sent to chatgpt. And then I'll say based on this, decide if you should buy, sell or do nothing. We can say in the assistant instructions, you are a world expert at stock trading, so let's ask ChatGPT right now, with an RSI of 58.87, what are you going to do? Test step. Here's Our output and the answer is nothing. A big fat nothing. After all of that, the real time trading agent decided to do nothing.

Okay, so far we've managed to get the trading signal from Taapi I.O. and we've sent it to ChatGPT to give us a result. Next up we want to add another step and this time we want to filter the step. Now in this filter we choose the field chatgpt and we're looking at the reply. Okay, now the condition, if the text exactly matches bye or the reply from ChatGPT exactly Matches sell, we will continue. Now you'll see in the case that I've set up, the result for the zap would not have continued because the response content was nothing. If the reply from ChatGPT was to buy or sell, as they'd say in the stock market, it will go on to the next step and buy or sell for us.

Next action is going to be a path and we'll simply call path a buy and path B sell. With that done, we'll go ahead and add alpaca in and we'll choose an event and we want to place an order. Let's go to continue. And we'll choose the symbol which of course, course will be Tesla. And then we'll choose a side which will be buy on this particular path and we'll choose a quantity of 2% of our portfolio. Well, that's our paper trading portfolio, $2,000. Next we do exactly the same on the sell action. We choose alpaca and we can see that everything is hunky dory.

So if we zoom to fit, you'll now see the full workflow that we've got right here inside Zapier. We've got the ability to retrieve the relative strength index from Taapi IO. We send it to ChatG, it decides whether to do nothing, buy or sell, and then it goes onto a filter. If you're buying or selling, continue the automation to do either buying or selling. If it comes back with nothing, then don't do anything.

Now this is a good fail safe as well because if ChatGPT replies in any other way apart from buy or sell, the automation stops. So no chance of any errors happening, which is great. But if we get a buy or a sell signal, the buy will go down a buy path. It will execute an order with alpaca and in the sell one it will do exactly the same thing. It will sell $2,000 worth of Tesla stocks.

I'm just going to set up the conditions for my paths. So again we're using ChatGPT's reply and if it exactly matches Buy, then it will go to the Buy automation right here. The same will go over to Sell and we'll do the same thing with ChatGPT. If the reply is exactly matching Sell, it's going to go on and execute here in the Sell action. Now you'll see the publish button goes blue, meaning we can publish this automation and start it when the market opens.

Now, if for whatever reason you don't want to use the baked in alpaca integration, you can use webhooks by Zapier as well. Enter all of this information, which of course I'll put in my community link will be down below and of course all your secret keys at the bottom and then you can do a test set step. So I'd love to test to see that this actually works before the market opens. Let's click test and boom. Yes, a request was sent. It looks like it's executed. And you can see here, yep, my buying power has gone down by $2,000. And there's also a recent order here that's just been placed. And of course this is before the market opens, but we can see when the market opens, we will be trading in real time with Tesla stocks. How exciting is that?

Okay, it's just about to turn 2:30 here in the UK. That means I can officially start my zap. Let's turn it on. And yes, we're up and running. We are trading in real time with Tesla stocks on the US stock market. You'll see if I refresh my AI trading agent, we have a success here. It's gone ahead and it's purchased some stocks in Tesla. Yes, I'm an official Tesla stockholder now, thanks to my automated agent.

And you'll see the conversation with ChatGPT, the RSI was around 37 and it decided the response was a buy. If we look at my paper trading account, you'll see yes indeed, I've spent some money and I've lost a little bit on that trade. But I am a proud owner of the Tesla asset here. I have 5.57 or so stocks in Tesla and they're currently going down in value, which is not great news, but of course that status was filled so my trade was executed and this is happening for me round the clock while the markets are open again. The same has happened. It's tried to sell Tesla stock twice here in this automation, but I haven't got enough stocks to sell because they've depreciated in value.

Now obviously if I was to make this way more advanced, I think in the future I'd add in a few more steps to my zap to tell the AI agent exactly how much money I've got to spend and how much I've got invested in the stock so it can make an intelligent decision based on all the information.

Now this is really cool. My stock agent is running. It's been running for a while and look at that. Oh, it's made $17. It just down. It was, it was $24amoment ago. This is exciting and nail biting stuff. $38. This is crazy. It seems that my AI trading agent is working exceedingly well, better than I could.

Now, as you can see here from the latest runs on the live market, a lot of the time it's been told not to do anything here and then it's just gone ahead and done another thing. Yes, it's purchased more Tesla stock. I really hope this works out for me. No, no, no. I'm going down. I've lost money. This is too much to handle. I'm going to go away of coffee and we'll see how my agent has done in just a few hours. I can't look at this anymore. The stocks have just dropped off a cliff. This graph here is not looking very healthy, showing you that there's the possibility for things to go very wrong just as much as there is the possibility for things to go right. Interesting.

Okay, and there we go. It's just after 9pm in the UK, 4pm Eastern Time. Trading has finished. So I'll turn this app off for now and take a look at my alpaca dashboard which reads out a total portfolio just gone up of $409.23 profit. That is pretty incredible. But you can see how things can go sideways as well. It was a flatline start leading to my nails being bitten as I lost nearly 500 when it spiked up and then spiked up and spiked up again to leave me over $400 in profit.

Now of course I was play paper trading. You could use this for real. But remember, it's your own responsibility and this video is not intended as financial advice. This is the price of a year subscription to Zapier Professional. So I'm going to give it away twice. That's right. I'm going to give away a free year of Zapier Professional to two lucky people. All you've got to do is comment down below and let me know how are you going to use this and what would you do with a Zapier professional account, how would you make your AI agent that trades on the stock market or in crypto? The best two comments will be replied to by me and winning a free year of Zapier. So get going now. And you never know, you could be the next person creating crazy profits on the stock market using AI automated trading.

Best of luck. Thank you to Zapier for sponsoring this video. And YouTube is showing a video on your screen right now. You should watch next.

Oh, hi there, Mike here from the future. So I've come back the next day and I've gained $2,446. I really feel like my AI trading agent knows its stuff. The shares are shooting up right now, and if we look down here, sell, sell, sell, buy, buy, buy. Essentially, my agent was working autonomously to trade on my behalf, and.