AgentZero: FREE AI Super Agent is INSANE🤯

agent zero. This free AI super agent is absolutely insane. So what I'm going to show you today is how I built for example this, and this all directly with agent zero. By the way, I've got a free ultimate cheat sheet inside the link comments and description if you want to learn more about this and what it does, etc. But basically with agent zero, this is a free AI super agent similar to Manis or Gen Spark or Plexi Labs and that sort of thing. And you can basically give it any prompt and it will go off and work its magic. So let's test out right now as an example of this. If we say okay research Julian Goldie and create an amazing landing page for his SEO agency with agent zero. This will basically analyze exactly what you want and then go off and do its magic. Right now at the same time you can actually have multiple tabs running. So if we start a new chat on a different tab we got the first task right here but the second task that we're going to run right here. So for example, we say okay, build a HTML synth keyboard. This will start running two different tasks at the same time using agent zero. Right? The cool thing with

this is it's running locally and it's really easy to set up using something called Docker. So if you download Docker, then it's pretty much just a copy and paste job where you can start running this locally in your terminal and then you're good to go on this and it's completely free. Now, if you want to get all the video notes from today whilst this is loading, you can get that inside the AI success lab link in the comments description. We've got the full setup guide on how to start using it, all the resource links as well, plus 100 different prompt ideas that we can test in a second to see how they perform. So, if we come back to this example, it said your HTML synth keyboard is ready. You can find it here. If we click on that, it will automatically download it. And then if we open up that file, we have the synth keyboard ready to go. How easy was that? Obviously, you're not going to be building synth keyboards all day. I don't think you are anyway. So, if we wanted to test something else out, let's see if it can connect to the internet.

Okay, we're going to say, okay, what CSMP 500 price today. Bear in mind, I'm running this locally. I would recommend running it in a virtual environment to be honest when you're building this out, just so you know, you can be safe with it. So you can run it inside Docker and also it's really easy to set up with Docker as well. So here we go. It said the most recent available closing price for the S& P 500 is 5,935 as of June the 1st. Realtime prices for June the 3rd are not yet published. So it can connect to the internet. It can look at things in real time. It can code. It can build out stuff. It can essentially look for different prices. Well, if we said, okay, plan out a trip to Tokyo here, you can basically automate and build anything that you want with one click, right? So, whether it's admin tasks like searching and planning out your neighborhoods or where you're going to be on that flight and the dates and that sort of thing, or whether it's, for example, building out HTML projects like you can see, it's really powerful doing this stuff. So, now, right, here's a 7-day trip plan for Tokyo for late September. Here's the flight prices, accommodation, transportation, top attractions, etc. And it's really fast to reply to you, right? As an example of this, we

wanted to compare it side by side. If we do exactly the same thing in a paid tool, for example, like Manis costs $39 per month on the normal plan. If we go over to Manis here, we'll test this out. We'll say, okay, plan out a trip to Tokyo. Hit enter. will do exactly the same thing inside agent zero at the same time. So this is in thinking mode whereas agent zero is already planning out trip it's already come back to us right so it's a lot faster and easier to use plus you can run it locally as you can see it's hosted on a local URL and then if you want to use this for free right so you can obviously you can set it up for free and all that sort of thing is fully open source but what you can also do is inside the API keys here you can just use a free API key from Google or you can plug in a free API key from open router, right? But either way, you can set up the super agent for free and you can also generate responses for free. So, it's so fast and easy to use. Let's test some other stuff out now. So

I'm going to go inside agent zero. We'll say create a one tap flying bird game with trippy backgrounds, evolving music, explosion effects, every tap triggers particle burst, blah blah. Right, single HDR. Let's see what we get back from that. So it's coding this out step by step. Basically what this is like a wrapper for an API, right? So you plug in your API key, but then it's connected to different tools. So for example, it can connect to like browser use tools or it can use for example response tools, right? And you'll see the tool name here. So for example, tool name equals response. If we scroll up, we've got the code execution tool. So, it's basically like taking the API you've got and then connecting it to loads of different tools and doing it in a no code way. So, if we have a look now, we've got the trippy game. As you can see, that was pretty easy to generate. You can always go back and forth in the chat as well if you want to improve this, but you can see how easy that is to build out. Now, if you want some ideas on prompts and that sort of thing, I've got 100 prompts ideas for agent zero inside the AI success lab along with the setup instructions and all the useful links and resources and some examples of what we built before previously. So, if we say now, okay, make a whack-a-ole game with sound effects. Let's test that out. And there we go. Got that ready to go. Let's click on that. Open it up. There we go. What's pretty crazy about this is just how quickly it works and how it pretty much works every single time. Like to me, that is wild because there's a lot of AI chat models and that sort of thing that actually don't work first time around at all. So, next up, what we're going to say is create a color palette generator. See what we got here. There we go. Super fast stuff. So, you can click on these and then it will generate different color palettes. Let's try something a bit more interesting now. So, I'm going to say inside here, create a drawing app where I can paint and draw on a whiteboard. Let's open this up. That works perfectly. This is crazy. There's also a bunch more settings inside here as well. So, if we go back now, you got external services. There's a development section as well, taskuler. So, you can actually manage, schedule tasks, and automate stuff. So, let's try that. I'm going to click on new task and we'll schedule this in. So I'm going to say okay do research on the latest AI news. You can choose whether you want it like ad hoc planned or scheduled. So you can actually schedule and then you set up the schedule system prompt user prompt and attachments. Right. So let's try this. And I'll just change this to ad hoc for now. And then you get a token to trigger this externally. So there's some pretty cool stuff there. You can schedule in tasks. You also got web browser as well. So what that means is it can browse the web, start checking stuff out for you live. There's a speech to text model inside here as well and utility model as

well. So basically so far we've covered how to build apps, how to build tools, games, landing pages using agency along with planning out trips and that sort of thing. There doesn't seem to be much it can't do to be honest with you. Let's try something else out. So we're going to say browse the web and take a screenshot of Google and the S& P 500 price right now. So, it's going to use the browser agent tool to start using this magic here. And we're going to go to finance. yahoo, open up Google, and take a screenshot. So, you can see it's actually using Google right here. It's actually taking a screenshot of Google. Now, it's navigated to Yahoo Finance, and it's taking a screenshot of the S& P 500 price. So, what's crazy about this is like it can just go off and browse the internet just like a human, but it's really fast when it does it. Now, when I'm using this model, I'm actually using GPT chat. You could use, for example, Gemini or whatever you want. And because it's all connected to different tools, it should be easy to browse the web, too. And then what you can see here is the progress. So, it's navigated to google. com. Then, we opened up a new tab in Yahoo Finance. From there, extracted all the information from the page, took some screenshots, and it can go through and do its magic right there. Now, also what you can do here is you can switch between the history of each tab. So, for example, if we want to go back to that trip to Tokyo, it's all labeled automatically right there. Really easy to switch through. Same with the synth keyboard and everything else will be built. Really cool stuff. I think as well you could probably give it login details and then it'll be able to log in and do stuff on your behalf too. But yeah, that's what let's try and put it to a new limit now. So, what I'm going to say is go to bageli. org or and generate an image of a cat. We'll see if it can actually do that. So, it's going to use the browser agent tool to do that. Now, you can see it's actually landed on the page from there. It's figured out, okay, it needs to click some buttons and move around, but it doesn't seem to be able to really like figure out, okay, here's exactly where to go. So, I'm going to give it the URL for the actual demo page and see if it can figure that out from there. Let's say use this instead. The good thing is you can interrupt it halfway through a task. So, that's pretty good. It started a new conversation, so it does seem to be able to use this tool. And then it's typed in generate an image of a cat on bagel down there. So, it's using the internet without us, which is crazy. And now we're in the queue and we can see that our image is processing right here. So we can actually use websites go off start speaking to AI directly as well. So this is our AI agent on agent zero and it's generating images inside Bagel which is another AI website. And you've got AI agents going off on the internet and then generating images autonomously without me. Right. Pretty crazy stuff. Let's see if he can navigate to YouTube now as well. I'm going to start a new chat and I say, "Okay, how many views did Julian Goldie SEO get on his latest YouTube video? " So, this time it's using the search engine and it's finding my YouTube channel. From there, it's going to use the browse agent to navigate to my YouTube channel. It's landed on the page, previewing the page, and it should be able to figure out how many views we got. You can see it's analyzing the images right here. So, it's got the thumbnails, it's got the content, etc. And then it's figured it out, right? So, it said, I found the latest video on Julian Gold's SEO YouTube channel. Video is titled I made blah blah. 721 views. This answers the user's question directly. Boom. There we go. All right. So, complete the task. navigated to YouTube, scanned the screen, figured out how many views we got, and you can see an example right there. It doesn't seem to be anything that's limited by. So, if we compare these outputs, right, so far, it managed to create an SEO landing page, which we can see right here. It created a synth keyboard, analyzed the stock market, planned a trip to Tokyo, created a drawing app, which actually works. went onto the internet and took some screenshots, generated an image via bagel. You can see the image is generated right there. It actually worked perfectly. And then also analyzed my YouTube channel and figured out how many views I got my latest video. There's nothing this agent can't do. It's insane how powerful it is and also how autonomous it is. And also the fact that it can use other AI engines like go on to bagel and generate images without us and doesn't seem to fail or anything. If you want to

get all the notes on exactly how to use Agent Zero, how it works, etc., feel free to get that inside the AI success lab. I've added useful links to everything that we've talked about today, some examples, 100 different prompt ideas for using this bad boy, and also resources in terms of loads of different courses on AI agents, AI super agents, etc., plus the setup instructions for using Agent Zero, like you can see right here. Now, if you want to get all my courses on like free AI super agents, how to use this stuff, all the latest recipes from this month, plus an awesome community with 613 members, you can ask for help, advice, and support. And also, if you want to learn about how to make money and save time with AI, plus get all my best workflows and AI agents workflows, feel free to get that inside the AI profit boardroom. On top of that, you can also post inside here and you can tell us what automations you want us to build. Um, each week we do a personalized video based on the requests here and just make all those automations or give you the resources to do them, right? And additionally, we have three weekly coaching calls. So, if you want to jump on live calls, ask us any questions, get help, get support, get coaching, this is the best place to do it. So, feel free to get that link in the comments description to the AI profit boardroom. And if you want us to just implement this stuff for you, feel free to book in a call, link in the comments description, and we can jump on an AI strategy session with you, look at what you're spending your time with, and then help you figure out how to automate that stuff. Link in the comments description. Appreciate you watching.