🚀 Get a FREE SEO strategy Session + Discount Now: https://go.juliangoldie.com/strategy-session
Want to get more customers, make more profit & save 100s of hours with AI? Join me in the AI Profit Boardroom: https://go.juliangoldie.com/ai-profit-boardroom
🤯 Want more money, traffic and sales from SEO? Join the SEO Elite Circle👇
https://go.juliangoldie.com/register
🤖 Need AI Automation Services? Book an AI Discovery Session Here: https://juliangoldieaiautomation.com/
Click below for FREE access to ✅ 50 FREE AI SEO TOOLS 🔥 200+ AI SEO Prompts! 📈 FREE AI SEO COMMUNITY with 2,000 SEOs ! 🚀 Free AI SEO Course 🏆 Plus TODAY's Video NOTES...
https://go.juliangoldie.com/chat-gpt-prompts
- Want a Custom GPT built? Order here: https://kwnyzkju.manus.space/
- Join our FREE AI SEO Accelerator here: https://www.facebook.com/groups/aiseomastermind
- Need consulting? Book a call with us here: https://link.juliangoldie.com/widget/bookings/seo-gameplanesov12
Unlock Free AI Automation with Google AI Agents and RU Code | Step-by-Step Guide
In this video, learn how to set up and use free Google AI agents through Visual Studio Code and RU Code extensions. Discover how to automate tasks like building apps, creating landing pages, and writing blogs using various free API keys, including those from Google and Brave Search. Follow the detailed instructions to get your API keys, configure settings, and set up MCP servers for seamless integration and task automation. Additionally, learn to switch between different AI models like Google Gemini, Anthropic's Claude, and Open Router to enhance performance and outputs. Join the AI Profit Boardroom for deeper insights, SOPs, and community support in automating and optimizing your workflow with AI.
00:00 Introduction to Free Google AI Agents
00:11 Setting Up Visual Studio Code and RU Code
00:57 Configuring Google Gemini API
01:56 Using MCP Servers with AI Agents
03:14 Creating and Automating Tasks with AI Agents
04:42 Advanced AI Agent Configurations
05:52 Building Apps and Tools with AI
12:51 Enhancing UI with Paid APIs
14:25 Exploring Alternative AI Models
21:17 Community and Additional Resources
Оглавление (10 сегментов)
Introduction to Free Google AI Agents
Today I'm going to show you some new one-click Google AI agents that absolutely wild and also completely free. So the way that you're going to set this up, I've got all the instructions inside the AI profit boardroom link in the comments
Setting Up Visual Studio Code and RU Code
description, but essentially you're going to go to Visual Studio Code. Make sure you download that. Then from here, you're going to go over to Rode, right? So you just go to extensions, type in R code, and then once we've done that, we'll see the icon for R code. So this is the icon, and you can see it on the left hand side over here. And from here we can start building out the AI agents with Google. So what we're going to do is we're going to go over to settings over here. And we can basically use these to for example automate almost very similar to manus AI. You can automate like MCP servers. You can build apps, games, tools, landing pages, write blogs, etc. And it is really simple and easy to do. So let me guide you through the process. What you're going to do from here is inside your settings configuration profile equals default. Then you're going to
Configuring Google Gemini API
select the API provider. It's Google Gemini and you're going to plug in your API key. Now, the API key is completely free. Let me show you how to get access to that. So, if you want to use this for free, you can just go to a studio. google. com. Then, you're going to grab an API key. Just create an API key over here and you can plug it into the Gemini API key section. Good to go. All right. Then, once you've done that, you can select which API you want to use. Now, these are all the different APIs from Google. Obviously, I believe if you use the later models or the thinking modes, that's going to rinse through more credits, which is going to slow down how much you can use this for free. If you want to use less credits, you can use like some of the old models. And the API allowance should be much bigger. All right, so we're going to select, for example, Gemini 2. 0 Flash. And also, you can see here I've auto approved everything. Auto approved writing, operations, etc. We can hit save. We can hit done. Boom. Shakalaka. We're good to go. All right. So, as an example of
Using MCP Servers with AI Agents
this, what we're going to do inside here is we're going to select code default and we can start using MCP servers to start interacting with the web so that we can connect this AI agent to the internet to do different tasks for us as a one-click sort of engine. Right? So, if we say install Brave Search MCP server, here's my API key. All right? And you might be saying, are the MCP servers free? Right? So, they actually are free. So, the way that you're going to be doing this is if you go over to Brave Search, Brave Search is free for the first 2,000 queries, right? So, when you sign up, you can install an MCP server that connects your AI agent inside Root Code and Visual Studio Code to get this set up, right? And it is free to do. All right. So, you can get 2,000 free queries like And then you're just going to get the MCP server set up like so. We'll get that running. Like you can see it now says the Brave Search MCP server has been installed and configured with the provided API key. You can now use the Brave Search tool to perform searches. So that means essentially what you can do now is you can use the free API key from Google. Brave Search API key which is free for 2,000 queries. And I don't think you're going to do more than 2,000 searches honestly. And then from here we can start using this AI agent
Creating and Automating Tasks with AI Agents
this oneclick agent to just go off and do stuff, right? So for example, if we say okay, search Google and find me the latest data about Manis AI. All right, it's going to use a Brave search MCP server like you can see right here. And then it's just going to start going off and start doing its magic. All right, it's going to run the server. So it needs to run it. It's not going to work like first time. So sometimes this can mess up a little bit. Now it started the server and we can start using this. All right. So, it doesn't when you're using the MCP server, just know that you have to go back and forth a little bit with this when you're doing it. But that's okay. Like, it is going to take a little bit of time. All right. And then if we go over to MCP servers, let's just check this is working. We'll just retry the connection right there. And we can start using that. All right. We've got another one set up here, for example, for Plexi. So, let's try that. We're going to go over to here and we're going to say, okay, use Plexi MCP server and search for information about Manis. And we just need to grab an API key from Plexi. This is actually a paid API key, but even then it's pretty cheap. But let's go over to settings. We'll grab an API key like so. Generate over here. Copy that. Plug it in over here. and this AI agent can just start going off and using Plexity. We just need to wait for it to set up the MCP server. If you're not sure what an MCP
Advanced AI Agent Configurations
server is, essentially what it does is it connects the AI inside R code to the tool that you're trying to use, right? And it doesn't need to interact directly with the interface. Now, there are ways to connect the AI inside R code directly to browser use. So, it can just go off and use the internet directly. But if you want to use it indirectly like this all inside R code, then you can do that. So for example, we've set up the perplexity ask MCP server like you see right here. Then we've run a query through it. We're using the Google agent API key. So we've set up the API key here so we can start using the Google AI agent. Then it's gone off searched about Manus using Plexity and you can see the task is completed. Right. So it says I've used the perplexity MCP server. Search for information about Manis. Manus is an advanced AI agent launched on March 6th, 2025 by the Chinese company Monica AI. Claims to be the world's first general purpose AI agent with a high degree of autonomy and intelligence. So it's really easy to set these up. Doesn't take long at all. And you can see how you can get this done in one click. Right? So now what we can actually do from here is we can take this to another level. So for example
Building Apps and Tools with AI
if we've used the brave search or the MCP servers from Publix, then we can start running tasks inside here and then using those to generate our apps, right? So for example, if we say research how and bear in mind like these APIs from Google, they're not connected to the internet unless you use the MCP server. So that's the gamechanging sort of way that you can use this Google or we'll say yeah research the latest update from Google Gemma 3 via Plexi and then create a landing page that shows how this will change the world. Make it sleek and beautifully designed in HTML. And then you can actually use the AI as well to improve the prompt. So if you click over here, we can now click on this to improve the prompt. And you can see we've got that set up ready to go. And it just makes the prompt much better. Right? So instead of like me being a lazy sort of human not typing this out properly, it will create a much nicer response like you see right here including like research across different healthcare and that sort of thing, right? Different industries, landing page instructions. It's a lot more detailed as a prompt. Plus, it understands, okay, this is what the AI needs to create something better. So, we're going to head into it's going to now use the MCP server from Plexi. We're going to prove that. And if you're wondering, okay, are the MCP servers free? All right, so for example, Plexi, this one is crazy cheap. Like, I've rinsed this and it's cost me like 60 cents over the last month. If you're using, for example, an MCP server, let me find these for you right now. That doesn't require an API key, then it's also free. So, for example, if we go over to model context protocol, then we'll find the list of maintained servers and then you can see all the different apps you can link this to. Now, if you're using an app that doesn't require a API key, then you can use them for free. So, for example, if we scrolling down here, like Airbnb, Airbnb doesn't require an API key to run the MCP server. So, if you can find MCP servers without an API key, then you're good to go. Or what you can also do is you can just find different MCP servers where the API key is free for a certain amount of tokens. For example, like Brave search. So let's go back. Right now we've got that set up. Like you can see we can now just run that command and it's done the research and created a page with the research about the AI. Right now that doesn't have a lot of information about it. So what I want to do now is go back to here and say okay using plexi get a lot more information about stats facts figures etc. Then use that to redesign the landing page and then we'll say make the landing page much sleeker and more modern in design. So now it's going to go back to perplexity. You can use this iterative process to just create whatever you want, right? You can go from idea to design in just a few clicks. You can also click here and always allow and then you know you don't need to approve that request every single time. So it's got a lot more research back now. You can see all the benchmarks that it got back from Gemma 3 and it's updating that inside the code for the page. It's also for example like search YouTube and other places like that. It's editing the files that we've got. It's also figured out some more technical features. And then if we run that command, we can start getting a better output from here. Right now, you can also use this to build like apps and stuff and tools as well. So, for example, if we go inside here, we'll start a new chat. And again, we're still using the Google Gemini API key, which is super cheap. And we're going to say, okay, build an SEO app for SEO cost calculators. And inside the settings here, I'm just going to change this to flash 001, which is the latest version, the latest API from Google. And now we can use this to start building out like apps and tools and that sort of thing as well. So wait for that to load. Hit save. Now it's editing CSS, then the JS. So, the JavaScript and again like I don't need to code to run this and then we've got the miniature app ready to go right there and it works perfectly nicely designed really simple really easy to use and you've just created an SEO cost calculator in the space of like literally 2 seconds it's crazy how easy this stuff is right what we could also do is we could say okay on the same page write a blog about SEO training underneath plus why it's important and write it Alex Hosi style. Now you see here how we've hit the quot on Google. That's because we're using one of the latest versions for the API. So if we switch the APIs now something a bit older like this. Hit save. Hit done. Go back. Cancel. Resume task. We can go back to using the API key. Right? So if you're running out of API credits or tokens, then you just switch the API key on the agent and then from there we can redo the whole thing. and just get going again, right? And so if you're ever struggling or if you're running out of tokens and that sort of thing, don't worry. Or don't think, oh, this is a way of Google trying to get me to pay for this stuff, you can just easily get the AI to switch API keys in the settings of R code. Change the API key over here and then you're good to go. All right. And then if we go inside the new version, so this was the old version and this is the new version. We've got the content right there for the page and you can just go back and forth inside the design as well. The other cool thing here is for example if we go to a website that you want to replicate for example like this one we can just take a screenshot and because Google has vision enabled we can go back into Visual Studio Code paste in that image that we created then say clone this and the AI agent can actually see the image that you created over here and then use that to start generating better outputs. So you see here how it's like redesigning the code and it's the user wants me to clone the design of the provided image. That's a significant change. Okay, here's what we need to do. And the cool thing about this as well is it's self iterating, right? So it's prompting itself. So literally all we did was say clone this and you can see how it goes back and forth in the chat with itself. Goes off, recodes the project, edits the CSS, edits the JS, everything else. And then we're good to go. Now I've updated it. I've improved it. Here you go. Here's a page and we've just redone the whole page. Right. The other cool thing as well is if you don't like the design, if you want a much better, for example
Enhancing UI with Paid APIs
front end, then you can go back in here and you can switch to something like Anthropic, which is a paid API, but it will give you much better outputs on the UI. Right? So, if you want to get uh much nicer UI, now we can go and use Claude 3. 7. We can switch over now and we can say for example, okay, redesign the UI. So, it looks much more modern, sleek, cool, and fun. All right. Now, so far the API cost has been zero. But now that we switched to Claw 3. 7 Sonet as the API key, which you can get from Anthropic, the UI is going to be much nicer. The code Obviously, it's going to cost a bit of money, but even then, it's going to be pretty cheap that you can build a whole website for. And you can see it's a bit slower as an API as well, right? So, this tends to be slower than Google Flash when it's coding and going back and forth of it, but it's also going to write better content like you can see. And now it's updating the file automatically. The cool thing about this as well is it's coding locally, right? So, if you ever want to come back to the files, you've got them all saved locally and you can just reopen Visual Studio Code, go inside recode, you're good to go on this. Now, if we hit save, it's going to recode the CSS. All right. So you can see it's cost us $18 to redo the HTML on this. So you can basically get Google to do all the fundamentals for free and then if you do want to add something a bit more special to the front end then you can switch to Claude and then just pay for the API key after that. Now this is not limited to
Exploring Alternative AI Models
websites or Google searches or using perplexity or Airbnb MCP servers. Whilst that's loading I'll tell you a little bit about it. So you can code for example games tools, content rises, etc. It's very similar to what Manis can do. But instead of for example waiting for the invitation code or waiting for the daily limit usage, you can use something like this. It's kind of like a free alternative to Manus, right? So Manis AI is a really powerful agent, but most people don't have access. It's not everyone has got an invitation code and also it's got daily limits, right? And it'll probably be a paid tool when it comes out. Whereas for example, you can use something like the Google API key that I've shown you today with root code and all of it's free. Visual Studio Code is free, the root API key from Google is free and then the MCP servers free if you use one without an API key or you use one within the limits. For example, like Brave Search. So let's wait for that to load. Some people asking like do you find that claw on it is slower than it? I've actually found the opposite. Like for example, you can see how much longer this task has taken with claw on it than it has with Google. Right? Google's just so much faster for getting the output whereas this is taking absolutely ages to redo the whole thing right and I know it's got to recode and improve the design but still it shouldn't take that long so tends to take longer from what I've heard 3. 5 set and 3. 7 set both from anthropic I think I saw someone comment on YouTube about that I haven't checked it out myself you never know with the YouTube comments all right so now it's updating the JS as well and also you might be wondering Okay. Can you use like open router as well? So you can actually plug in open router as well directly into root code. I'll show you that in a second once this is finished. So once this is finished editing, I can show you some examples of other free API keys you can use so that if you don't want to use Google then you don't have to. So let's see what we got here. We'll run the command. We'll open this up. And this is totally different. Let me show you how different this is. All right. So you can see now you got the front end from Claude right here. And it's just it looks totally different to what we had before. All right. So here was the output from Google. It did the research did like the most of the research there, but it's just nowhere near as good as like the style from 3. 7 Sonic. Right? This is like a website from 10 years ago. that has some really cool images here. PreSEO strategy section top right. looks really modern. The links actually work. The tool looks a lot more modern and you're good to go on that. Right? So, that's the difference between these tools. And I think the Claw 3. 7 Sonnet, bear in mind, let's see how much that cost us. That cost us50 just to redo the front end, whereas a normal freelancer, you're paying like anywhere between $100 to $1,000 to retouch the front end design. And you can see the difference here. It's just 10 times better with one single click. So, really powerful. And then yeah, like I was saying, if you want to use other free AI agents inside root code, select open router and then we can switch between them. So let's see what we've got here. Okay, so we need to switch down here and we if you just type in free. So if we delete that, type in free here, we can switch between all the different models. So for example, you got Deepseek, R1 Distilled. You do have other versions of Google as well. What else we got here? We have hugging face meta llama. Mistral could be really good. This is a new model that was outperforming Gemma recently. So Mistral 3. 1 could be a good option. Let's test it out and see if that's good. Quen I really like as well. Quen is a great model for coding out. So let's have a look what we got here. Let's try for example Mistral 3. 1 free 24b. All right, we'll click on that. We'll hit save. We hit done. It's off that. We'll open a new task. and we'll say okay create a self playing snake game and if you're wondering okay how do you get the open router API key so you just go to openrooout. ai AI. Then from here, you're going to log in, go to keys, and then inside the key section, just create a key here, and you can use that to start coding stuff out. So, that's actually working as well, like you can see. And this is more like a thinking model, right? So, you see here, I was like, I need to create stuff playing Snake Game, etc. It's a really small model that's just come out from Misro. So, Misro 3. 1. I actually have a video on YouTube if you want to see all the benchmarks and how it performs. But it was actually outperforming in a lot of ways Gemma 3, which is the latest model from Google. So, pretty cool stuff. And you might be looking at that website as well before and thinking, where is it? Where's that website gone? That website that you saw before anyway, the front end that was done like it is better than most developers can do, let's be honest. And like, how long would it take to learn how to code something like that? like he's absolutely crazy, right? So, like for example, if we go back to this, we can just reveal and find her. Here you go. And then we'll open this up on a new Chrome tab. Plug that in. Yeah, we got the snake game ready to go. But where is the I don't want the snake game. Oh, it's disappeared. I wanted to show you the cost calculator again, but you saw it before earlier anyway. Like the front end was so nice. Like it is a genuine threat to front-end developers and that sort of thing. Plus, it doesn't take years to learn how to code and it's just super smart already. All right. Now, some people are asking like what other models actually allow computer use. So, if we go in here and we go to open router, you can see that this doesn't support images and it doesn't support computer use. I don't know many other models actually allow computer use. You can, for example, switch to I think chat GBT might be up to do it. I'm not even sure as far as I know. The only one that I've really tested for computer use is claude. So computer use for anyone watching if you're confused and you're wondering okay what AI can use computer use. So it looks open AI have an option and then also claude but the only one that I've really tested out is claude chat GPT. They have chat GPT operator but honestly it's not as good. You can actually get better options inside convergence AI and convergence AI is free. All right, let's see what else we've got here. You've also got a human relay which is another option but I'll come on to that later. So that's basically it. Like this AI agent is absolutely insane. You can basically automate anything with it. It can create very similar things to Manis. So it's a great free alternative to Manis and the API keys are free. The MCP servers I've shown you how to do for free. Root Code is free. Visual Studio Code is free as
Community and Additional Resources
well. If you want to learn how to set all of this up properly, I have a SOP inside the AI profit boardroom. If you want to learn exactly how to set this up for right, so if you want all the links, all the step-by-step tutorials, etc., feel free to get that inside the link in the description to the AI profit boardroom. This is a community basically focused on making more money and saving time and automating with AI, right? And if you go to the SAP section over here, we've got all these SAPs on a manus AI course. We have a deep seat course in there. This gets updated pretty much every single day. Comes with my best MCP agents and N MCP agent tutorials as well as an SAP for root code and computer use setup as well. And also if you have any questions, if you get stuck, etc. You can post inside the community and get help right there. Right? So you can see for example everyone, it's a really active community, lots of people asking questions in there and they get answers really quickly. So feel free to get that. And also this comes with weekly Q& A with us. So, if you have any questions, you can just jump on the weekly Q& A and ask us any questions that you have. Additionally, if you want a free one-to-one SEO strategy session, feel free to get that link in the comments in description, and we'll show you how we take websites from zero to 145,000 visits month and generate hundreds of thousands of dollars in sales multiplier. On this free SEO domination plan, you will get a custom tailored link building plan to generate more leads, sales, and profits from your website. Plus, you'll be able to answer any questions you have, and you'll learn how to 10x SEO traffic based on what's working for us and a happy gliders like you can see right here.