I Built an AI That Automates My Morning Routine (Using n8n)

What if you can wake up every single morning to an email digest that is perfectly dialed in and tailored to exactly the information and news that you want to hear about? Doesn't matter what niche you're in, or if it's a super obscure topic that nobody else cares about, you can actually create an AI agent that will send you an email digest every single morning. And that is just scratching the surface of what you can do with a tool like N8N. By the end of this video, you'll be able to build an automation that collects news from all over the Internet, no matter how obscure this source is, and sends you that daily digest? However, this is just scratching the surface of what an automation tool like N8N can do. You can literally automate anything. You can have N8N turn on your lights when you pull into your garage at home. Fireship built an automation where when his wife texts him with some sort of mad text, he automatically sends her flowers and tweets an apology for her. I mean, really, what you can do with a tool like N8N is endless. But in this video, I want to make a fairly simple demo to really show you the possibilities and how to build one of these workflows. But the idea is really to get your wheels turning on what else you can build with this, because quite honestly, the opportunities are pretty endless. Now, when it comes to using N8N, there's a few ways that you can do it. First, you can go over to N8N IO, which is the official N8N website, and you can use it to a limited degree for free. But beyond that, you're looking at a minimum of 24 bucks a month to use N8N's direct website. However, N8N is openly available. You can actually find it over on GitHub with all of the source code available. So you could actually install this locally as well and run it completely for free if you want. In fact, if I open up my computer's terminal and just simply type NPX N8N and hit enter, it asks me if I want to install N8N locally. I could hit Y, it'll go through a bunch of installation steps and then it tells me to press O to open in browser. So let's go ahead and press O. It's open in a browser tab. Here you can see, because it's on localhost and I can go ahead and create an account, which I'm gonna do real quick. Just like that, I can start creating workflows. We can see. Here's our N8N automation builder and our goal is to get a final automation that looks like this. There are however, some downsides to running this locally. One, it can be complex and overwhelming if you're not comfortable using the terminal to set something up like this. But more importantly, if you're running it local, that local server needs to be running for to run any of the automations inside of N8N. So if your Internet is ever down or your computer is turned off, you're not going to be able to run your N8N installation. Or let's say you're traveling. I'm in hotel rooms a lot, going to various conferences and keynotes and things. If one of my automations isn't working properly, I have no way of accessing it on the road because it's only installed on my local computer. So my recommendation isn't actually using the N8N website, it's not installing it locally, it's installing it on a third party Linux server that's hosted in the cloud that you can access anytime. And for that I actually recommend Hostinger, which is the sponsor of today's video. So we're going to build our email newsletter automation here in a moment. But first let's create a hostinger virtual private server to run this on so it constantly runs even if my computer's turned off. If you head over to hostinger.com and/Matt N8N, you'll land on a special N8N focused page for signing up to a Hostinger account. I recommend the KVM2 plan which starts as low as 6.99amonth and you can actually get it cheaper. Let's go ahead and choose the plan here. And if I come over here to have coupon code and I enter the coupon code Matt Wolf, click Apply, it'll get even cheaper. Pick the location that has the best latency for you. And then when we get to this page where it asks you to choose an operating system, well, we want this to be purpose BU for N8N. So I'm going to click on application, we'll click on show more and then if we scroll down to developer tools, you'll see N8N right here even has a star next to it to show that this is the popular one that you want. So we'll click that, we'll pick a password, it'll go through a quick setup process and then we're in the back end of Hostinger. Now all I have to do is click on Manage app and we get a similar page where it asks us to create our account for logging in. So I'll go ahead and do that real quick. And we are in N8N now to build whatever we want. Now the first thing I recommend doing before you actually go and create your own workflow, which we will do in a second, is to check out all of the templates that are available. There's almost 6,000 workflow automation templates available for N8N that people have already just built the automations. You just need to go and swap out the details with your specific details. There's pre built Automations to generate AI viral videos with VO3 and automatically upload them to TikTok. You could analyze a landing page with OpenAI and get optimization tips. There's nutrition trackers, automated economic calendars, stock trading signals with Gemini 2.5. So much more. I mean again, almost 6,000 templates, but let's go ahead and build one from scratch and really show off what N8N can do. So inside of our N8N dashboard, we'll go ahead and start from scratch and we have our node builder here where we can start building. And you need your first step to be a trigger. What's going to make this automation actually start? So we'll add our first step and we'll set this on a schedule because we're building a daily newsletter so we want it to send every morning trigger interval, we'll leave as days, days between trigger we'll use one trigger at our let's have it go at 7am that way it's waiting for me when I finally sit down at my computer in the morning. We'll click on Execute step and we can see it executed properly. So let's go back to our canvas by clicking the top left link here. And we can see we've got our schedule trigger to schedule every morning at 7am Next up, we want to give it a set of sources to pull the news from. So we'll add our next step. I'll click the plus here we'll click into data transformation and then we'll use edit fields, which also is a way to set fields. So we'll select this one. Now this step might get a little confusing, but if you follow along, you'll be just fine. Under edit fields here, when we get this new screen, we're going to click add field and then for name we'll just call it rss. These are going to be the different RSS feeds where we pull news from and then instead of string we're going to give it an array, because we're going to give it a list of various URLs to pull news from. Then under value, I'm going to click this little button to make it bigger so I can actually type a little bit easier. And we're going to enter just some very, very simple JSON. I'm just going to put this open bracket here, go down a line, I'm going to put a quote, and then our first URL that we want it to pull from is going to go inside of the first set of quotes. Now we need to find the RSS feed from websites. This is a feed where the websites automatically send this news out to other websites that aggregate this. So let's say I want to find it for TechCrunch. If I go to TechCrunch.com, we need to find the RSS feed and us. Essentially it's just a matter of adding like RSS XML to the end of the URL. If that doesn't work, then sometimes you just need to add feed to the end of the URL. And in this case, we found the rss feed for TechCrunch. I could grab this URL here, copy it, jump back to N8N, enter this as our first URL, and then we're going to go ahead and close off the quotations and put a comma, and then we're going to do it for our next URL. This time, let's go to Engadget.com and I believe this one might use RSS XML. And here's our RSS feed for this one. So I'll copy this feed here, we'll plug it in, close our quotation, put a comma, and then we'll add some more. So I added a couple more. Keep in mind this last URL does not need a comma at the end. We just need to put another line break and then close off our brackets here. And this is our simple little list of URLs that is going to pull our RSS feeds from. We could go ahead and close out of this click Execute step and we can see our output shows our four URLs. So we know this step is working. Go back to our canvas. And now we can see our first two steps. But now we wanted to look at each of those feeds individually, so we wanted to go through one feed at a time and find the news from each of them. So we'll actually go ahead and split out each URL one by one. I can add a plus here. We'll Click on Data Transformation again. And under here we can see an option to Split out. So we'll select Split out fields. To split out, we'll select the RSS field here. We'll just drag and drop it right there. And now it knows it's splitting out the RSS field. We'll click on Execute Step and you can see it broke out each of these four fields into separate fields. Now we need it to read the RSS feed. So I don't know exactly where the RSS node is in the subfolders here. So I'm just going to search RSS here. And you can see we've got RSS read. So we'll go ahead and select RSS Read. So it reads that feed there. And then under URL we're going to pull over this RSS that came out of the split out. So if I grab this, drag it in, you'll see it added our little code here for JSON rss. So it's going to read the feeds of every URL that we gave in the step two steps ago. We'll click Execute Step to see what happens. This one's going to take a moment because it's actually reading through all of those feeds and a few seconds later we can see it output 130 items here. 13 pages of 10 pieces of news per page. And we can see exactly what information it's pulling from the RSS feed. We could see who created it. We could see the title of the article, the URL, the publication date, the exact content, the content snippet, the categories or tags associated with the post. So lots and lots of information that we can now pull from those various feeds. So let's go back to our canvas and the next thing I want to do is I want to take all of that news, all 130 articles I just found, and put it all into just one single write up. And we're going to have AI summarize all that for us. So it's going to take all of that information. I'm going to click plus here and I'm going to do a quick search for aggregate and we'll use the aggregate node and then we'll leave this set as individual fields. But for our input field name, let's pull the content snippet. The content snippet is just this sort of like one to two sentence summary of what that news article is about. So ideally there's enough in there for us to work with that information. So we'll go ahead and grab content snippet Pull that into our input field name. Here we'll click Execute this step. And now look at this. We have one giant piece of content here that I can just scroll forever and ever and ever. And it's got all of those new snippets from all of those articles that we just came across, about 130 of them. Here's where it gets really fun. Now we're going to pull AI into the mix. So far this has just been sort of scraping data from around the Internet for us, but now we want to clean it up and give us just what we care about. So we'll go back to our canvas, we'll click on plus, and this time we're going to click into the AI box and we're going to select AI Agent. Once we're in the AI agent, we're going to set up a few settings here under Source for prompt. We're going to set define below. Because we're not actually entering any sort of text prompt on a daily basis, we want it to use the same prompt every time. For our prompt here, I'm just going to type Summarize news because we're going to actually put most of the details that we want into the system message. So I'm going to select add option. Here we'll select system message and this is where we enter a more detailed prompt for it to work with. So let's give it a prompt like only grab the news from the past 24 hours, act as a tech news aggregator and write in plain, easy to understand English with no tech jargon. Here is the list of news. And then down below that, I can pull in our aggregated content snippets. So you can see here's all of the snippets that it pulled in. But I don't want it to pull it in in that format. I want it to pull in all of the content snippets. So if I minimize this and just tell it to pull in all of the content snippets, you can see it pulls in just that JSON there. And it'll assign it with this prompt. So it'll tell it to only grab the news from the past 24 hours. Here's a list of news. And then it'll pull in all of that content that it just scraped from all of those websites. Now if I want to see what this prompt actually looks like, I can click on this little box down here to expand our system message. We can see our system message here, but we can also see the result that it gave. So here's our prompt, here's the list of news, and then here is that dump of all of the news here. We'll close out of this and it won't work yet because it says a chat model sub node must be connected. So we need to connect this to a large language model to actually have it use this prompt that we gave it. So down on the bottom, if I click on add chat model down here, we can select any model we want for this. We'll keep it simple. Let's go ahead and use an OpenAI chat model and we will have to share an API key. So I'll go ahead and create new credential and it's going to ask for my API key from OpenAI. Now, in order to get this, head on over to platform.OpenAI.com, click on Dashboard in the top of the screen and then on the left click on API keys and create a new secret key. I'm going to call this one N8N. It's going to create a secret that I can go ahead and copy here. Now when I jump back to N8N, I could paste in my API key right here. Now simply save and we're connected to AI. Now let's go ahead and pick a model and you can pick from any of the OpenAI models that are here. Let's go ahead and use GPT5. Now if I go back to my canvas here, double click into our AI agent and execute this step, we'll actually see what the output will look like now. And we can see this is the output that it gives us. So it took took all of this content that we could see on the left, tons and tons and tons of content, and turned it into this. The problem is this won't look very friendly in my email. This is going to look exactly like we're seeing it now when I send it over to email because it's using a lot of like Markdown language and email reads and HTML. So let's go back to our canvas and let's go ahead and tweak the formatting so it's something we can read in our email inbox without making our heads explode. So I'll click the plus here. And we want to convert Markdown to HTML. So I'm going to go ahead and search for Markdown and we can see we've got convert data between Markdown and HTML, exactly what we're looking for. We need to switch this mode from HTML to Markdown to Markdown to HTML because we're starting with Markdown. The markdown that we're going to be converting is this output. So all I have to do is click output right here and drag it into Markdown. Let's execute the step to make sure it's working and check that out. Now we've got it in HTML. Here's a quick summary of the top tech news from the past 24 hours, all broken up into HTML. Now we just need it to email us. So we'll add our final step here and I'm going to do a quick search for email and then we'll go ahead and select Gmail as the option and then for actions we will send a message. Now, quite honestly, this is probably the most involved part of this whole process is actually getting your Gmail account set up inside of N8N up here where it says Credential to connect with. We need to create a new credential to connect it to Google Google. So I'll click Create new Credential, but then it asks for our OAuth redirect URL, client ID, client secret, all of this kind of stuff. And well, we gotta find all of that information. And there's a handful of steps you need to follow. Luckily, if I click on Open Docs here, we can see we have explanations on exactly how to set this up. So once we're on this screen, we pretty much just need to follow these steps. We need to create a Google Cloud console project, enable APIs, configure OAuth consent screen, create a Google OAuth client credentials and finish the N8N credential. And again, this will probably take you five or six minutes to go through all these steps. But if you follow these steps exactly, you will get the credentials. Now, during this process, there's one step that might trip people up and that's the step where it asks you to add an authorized domain. When you get to this step, click on Add domain and then for the authorized domain, if you are hosting it over on Hostinger, the authorized domain is going to be this domain here. So we'll go ahead and copy that and we'll paste this into our authorized domain. Again, if you follow all the steps on the documentation page, you should get set up fairly easily. Now we simply just need to dial in the email that it's going to send. So we're connected to our Gmail account that we just set up. We'll leave the resources message operation, we'll leave as send. And while I want this to email me every day so let's go ahead and put in MattutureTools IO for our subject. We'll call it daily the Tech Digest email type we'll leave as HTML. And then for our message you can see we've got our original which is in markdown. And then we've got our data here which is in HTML. We want it to pull it in in HTML. So let's go ahead and grab this box, pull it into our message. Now, if I click Execute step, you can see that it says it sent the email. Now it doesn't show the output of the email because this step is just sending an email. So the output is email was sent. But if I go check my inbox, you can see I just got an email from, well, myself that says Daily Tech Digest. If I click on this, here's the email, perfectly formatted and everything. Here's a quick summary of the top tech news from the past 24 hours. AI and regulation, big tech and legal, various gadgets, new devices and updates, startups and ventures, cybersecurity. Now, I didn't dial this in just to be AI, this is all tech news. But I just built an entire automation where every Single morning at 7am it's going to send me a tech digest of the last 24 hours. And again, this is just scratching the surface of what's possible with N8N. If we wanted, we can have this tech digest sent directly to our Slack channel or emailed to an entire newsletter through Mailchimp or ConvertKit. The possibility for automations and agents that you can build with this thing is just insane. There is so much you can do but to give the quick wrap up of everything that's happening in this automation. So we start with a trigger. This trigger fires off every day at 7am it looks at these news resources here that we plugged into our little mini JSON. It splits those URLs out so that it can look at it individually. It then looks at the RSS feed of every single One of those URLs that we gave it aggregates it all into one giant message. That one giant message goes into our AI agent with a prompt to chat GPT which summarizes just the important news from the last 24 hours. The output that it gives us is then converted from markdown into HTML so it reads better as an email. And then finally it's sent to us in Gmail as a daily digest. And this will run every single morning at 7am really, really powerful workflows that you can build again just Scratching the surface. This is designed to get your wheels turning about what's possible with N8N and honestly, the easiest and best way to get it set up and running in a secure safe always on way is to build it using Hostinger. So thanks again to Hostinger for sponsoring this video. You can go directly to their N8N integration page by going to hostinger.com mattnate n. And then when you're checking out, don't forget to use the coupon code Matt Wolf to get an extra 10% off. And also, if you want to read a newsletter about AI that's still curated and written by a human, make sure you check out FeatureTools IE and join that newsletter. I send an email every Wednesday and Friday with the latest AI news and tools and well, well yeah, that one actually is still written by a human. You can find that over at FutureTools IO. And one final thing before I wrap up, just a little bit of housekeeping. I am giving away an iPhone 17. I just got one myself. The camera on this thing is absolutely insane and you can get one for free. Super easy to do. All you have to do is go follow me on Instagram and I have a post mentioning the iPhone giveaway. Follow me on Instagram, comment on that post and you're entered to win. I'm really trying to grow my Instagram account right now and I want to grow it so badly that I'm willing to give away iPhone to try to do it. So head on over to my Instagram, follow me and comment on the post about the iPhone and you'll get entered to win as well. Thank you so much again to Hostinger for sponsoring this one. I am so blown away by what you can do with N8N and these various AI agent builders that are out there. I know my wheels are really, really turning about the types of automations you can build with this stuff and hopefully yours are too. Thanks again for tuning in, hanging out, nerding out with me. Really, really appreciate you. Hopefully I'll see you in the next one. Bye bye. Thank you so much for nerding out with me today. If you like videos like this, make sure to give it a thumbs up and subscribe to this channel. I'll make sure more videos like this show up in your YouTube feed. And if you haven't already, check out FutureTools IE where I share all the coolest AI tools and all the latest AI news and there's an awesome free newsletter. Thanks again, really appreciate you. See you in the next one.