How I Automated Faceless Shorts with AI (FREE N8N Template) 🤯

Today I'm going to show you how to automate shorts like you can see right here. So this is an example of a short actually automated with NA10. And it's actually not that hard to automate this. I'm going to give you the full free template today as well as I'll show you how to post plus create the video and you don't even need to come up with the idea yourself. Now it's pretty simple. So here is the full automation that I used to create it. So you can see an example right here. This is the example of the short that was created and then the final output. And then additionally, these are the accounts that can go viral. So you can see a bunch of AI automated animal related videos absolutely popping up on Tik Tok and you can now easily automate them using the process I'm about to show you. So let's

get straight into it and I'll show you exactly how this works. So essentially what we've got here is stage one, stage two, stage three, stage four, stage five, and stage six of the process. This is actually an automation I learned from Robbo Nuggets on YouTube. You can see his original video right here. Would definitely recommend checking out. But essentially, this is how you can build the full automation. And if you want to get the full template from me, you can get it for free inside the AI success lab. Just click the link in the description to get access to that. You can download the JSON then import it directly into NA10 as a free template.

So if we have a look at this automation, basically what we're doing step by step is first of all we are generating the idea. So you can see here we've generated the idea and this is the idea generator. We're using chat GBT basically to just come up with related and relevant video ideas that we can create and automate a short about. Right? So you don't even need to come up with your own ideas. If you do have something in mind, then you can easily put that into the prompt. And let me show you exactly how this looks step by step if you want to build it yourself. So, we're going to open up a new tab inside NA10 if you want to have a look at this. I've actually got a freehour course on NA10 on YouTube if you need to check out anything more on this. And essentially, if you go over here, open up the workspace, you can start creating your own workflow just like this. If we go to create workflow in the top right, and then from here, we're going to start building stuff out. Now, the cool thing about this is it's automated. So, it's scheduled to trigger every hour if we want to. Or we can trigger midnight or whatever we want, right? So, we can have days between triggers. We could set this up to trigger every hour. So, we could say every 60 minutes. Generate a new short. And it's incredibly easy to do. So, first thing you do is just set up a trigger. So, let's say, for example, schedule trigger. And if we switch this now to hours, and let's say you want a short to come out every 6 minutes. You can set up hours then 1 hour and then you're good to go on that. Right from here and this is stage one of the process. We're then going to come up with the idea generator like you can see. So the way that we're going to do this is click on the plus sign and then from here we shall click on chat and open AI and then the method for doing this is message model resource equals text operation equals message model and model is I honestly for speed I would recommend using chat GBT40 but if you want like the best quality prompts then I would recommend using GPT 4. 1. So that's up to you how you want to set this up. But if you want to get similar outputs to what I created, then you can easily use chat GBT4. And obviously that's going to be a lot cheaper for you as well. Right inside here, we can simplify the output. And then we're going to put the output content as JSON. And then inside this section, we're good to go. All right. Now, we just need to select this prompt. If you want the full prompt or if you just want to steal this stuff from me, then just get it inside the AI success lab. You can grab the blueprint down here and you can see all the resource links like but essentially if we paste that prompt in directly here this is going to come up with the ideas for our prompt for our videos. So you see here it says you're an AI designed to generate compelling viral video short ideas for social media. And I've put just for fun I've put make it jungle themed about animals in the jungle. You could easily change this and say okay make it SEO themed for example whatever your niche is whatever you want to post about it doesn't have to be about anything in particular right so whether you want it to come up with ideas for you whether you want the ideas to be generated automatically you can do it directly inside this prompt here and if you look at the outputs like it's pretty amazing what it came up with so for this example the caption so it's like monkey steals tourist GoPro and films jungle life it came up with the hash hashtags, posted it automatically to social media. And then you can see this is 1 minute long, this short, which is pretty cool as well. You could also monetize the account and the images are super nice. In fact, these are not images, these videos, but you see the point. And essentially, then the monkey just goes off and films the jungle and that sort of thing, right? Films always chunky mate, jungle maze. And so this is basically how it goes step by step. And then from here, what you're going to do is you're just going to change this. So, let's say for example, we're going to say make it SEO themed and educational, right? Then we're just going to test this step. At each stage, you want to test the step, make sure it actually works because if you don't test the steps, then you don't know if it's going to work or not, right? So for example here for our business related content it said the day Google got a cold comes up with the ideas and then it's like a quirky animated story follows Google the Google robot who catches a cold and accidentally buries a small business site until an SEO detective Metamax saves the date with metadata and backlinks. Right? So it's very creative in the way that it comes up with ideas and it's all automated using NAN. So once you've set this up, you've now got the idea generator, right? And that is stage one. Stage one is complete, my friends. And that was really easy, right? Now stage

two is actually taking the video prompts. So taking the ideas and then generating video prompts. Now, if you want an example of this and you want to know how to set it up and just set it up nice and clean like you can see, so you got stage one, stage two, etc., then that's pretty simple as well. And so all you do is you can press shift and S and that will generate a sticky note for you and then you can annotate that. All right. So for example here we'll put stage one idea generation. We're good to go. All right. So stage one idea generating. Here's a schedule trigger. Here it is step by step. Boom. Now we're going to do stage two. And I'm just going to resize that so it looks nice and neat. And here we're going to duplicate and take this down. And this will be stage two. We shall call this idea to prompts. So now if you want to change the color of the sticky note, just keep it nice and organized. You can do it like so. So you can select different colors right here. And then we can keep going along. So the next stage is to take the idea and plug it into a Google sheet. And this just keeps everything organized. Plus you can see what's ready for production, etc. So if we type in Google Sheets over here then from this section we shall select if we click on this just double check sheet within document. All right so the operation is going to be append row. We're going to select append row and then I'm just going to choose a spreadsheet from my Google drive. Let's take that one. Then we're going to take the sheet here second. Here we go. and we'll take the sheet and then we're going to map each column manually inside the spreadsheet and essentially just so you understand what's going on here. So at this point we are creating a nice Google sheet template like you can see where we've got the idea the caption the production environment and the final output which is actually the video file right so if we click on that actually take us to the video file to download we've got the video file downloaded and ready to post to social media and you can actually automate the posting of social media as well which I'll come on to so you can see the ID the caption production environment final output and the ID as well what you would do is you would just map out the document. By the way, this is inside the school group. You can just grab the NATM workflow and then make a copy of it. So, you just go to file, make copy of the spreadsheet. And from here, the ID we're going to put row minus one. All right, that's basically going to label the ID. So, video one, video two, etc. Minus one because obviously we're not starting from row one. We're starting from row two. Row one has the headers inside it. Then from here we're going to take the idea which is generated by chat GPT. We then have the title, the environment etc. All right. So the caption we can take as the title. The production we can just type in like for production for example. Environment we can just drag and drop that. And that's going to be dynamic each time this workflow generates. And then final output we keep blank cuz we're going to generate that later with VRL which I'll come on to in a minute. All right. So this is basically how you can map out the content of your videos into an NA10 workflow. Pretty simple and easy. So we got stage one idea generation. Stage two idea at the prompts. And then what we're going to do from here is we're going to set up the prompt generator. All right. So this is a message model. We'll click on plus. We'll type in message. Type in chat GPT. Then message model. And if you want to save time here, you can just duplicate this, right? So, we can right click this, click on duplicate, and then from here, we've got the AI model duplicated, right? Just saves you time when you're setting this up. Now, you can also rename each module just to keep it clean and organized. So, for example, here we can put video idea generator. And then inside this section, we'll call that prompt generator. And if you have any questions as we go along, feel free to comment. Happy to help or post inside the AI community link in the comments and we can help you out. All right, so with the idea generator, then we're going to do the prompt generator, which we got here, and we're going to start setting this up. Now, the way that we're going to set this up is we're going to just copy the original prompt. So, we're going to open this up and take that and plug this in. Right. So basically what we're doing here is we're saying you're a hyper realistic prompt engine built for generating ultra specific documentary style image prompts. Make the prompt simple and easy. Okay. And then your objective is to create a 13 scene static photo sequence. Here's what to do. Here's how to do it, etc. And the first thing we need to do just before we do that is just test each step. So we're taking the data from the previous module, plugging it into the next one. And also if you want to save the outputs. So let's say for example you don't want to generate this every single time. You want to pick up where you left off. You don't want to have to test each part of the workflow each time. Then what you can do here is you can actually pin these. Right? So if you right click the module, click pin, this will save the output. So you don't need to test the workflow every single time you're setting this up. All right. And then you can also save here as well. So, inside the prompt generator, I'm just going to take this information, plug it in here, then we'll click test. And there we go. So, you got examples here of what we're doing step by step. Now, we're just going to flick to expression like you can see. And inside expression, basically, this allows dynamic fields. So, it takes the inputs and the ideas and the environment and everything else and plugs it over here. So you always want to be testing each step just making sure that the outputs are what you expect because if you don't get the right outputs from the previous module the next module is not going to work very well. So you can see here now it's come back with the ideas the stepby-step scenes etc. So Google Chrome Roma blah blah here's what it's doing step by step and there we go. And by the way if you just want to copy and paste all the prompts for me get it inside the AIC success lab. you can just steal the JSON right there. Right. And then from there, you would just go to import from URL or import from file and upload the JSON file. So you can see here we've got some formatting. Now this just makes the code more accessible for the next steps in the process. So we're going to click on add module, click code, and then we're going to take the code from here, plug it in, test out a step. And now you see how this has turned all the stepbystep scenes from here. So we've got scene one, scene two, scene three, etc. It's plugged it into a full description here, which is going to make it better and easier for the next stage in the process. Right? So when we're asking the AI to generate the outputs, this is going to be much easier digest for the AI. And that is stage two completed. All right. So stage one, idea

generation. Stage two, video prompts. And then we're going to move on to stage three, which is generating the images. So, I'm just going to duplicate this sticky note again. We'll plug that down here. We'll rename this. So, this is where it starts getting interesting because now we're going to generate the images, right? So, even if you didn't want to make videos, you can now generate images on autopilot, too, like you can see. So inside the generate images section, what we're going to do now is a HTTP request, right? And the model that we're going to be using for this is Flux. Now Flux is pretty good for generating images in a cost effective way. And the way that you're going to do this is click the plus sign down here and then you're going to type in HTTP quest. switch this over to post. So we'll take this and then the other thing that we need to do here is select send headers. So if we select send headers then we're going to take authorization and we're going to get an API key which I'll show you how to do in a second. All right. So if you want to get an API key you're going to go to F AI. Now F AI is really good for generating videos, sound, images, etc. And it's all inside one single platform which makes it much easier. It's kind of like open router but for multimedia, not just for generating text. So with FA, you can grab an API key. You can see I've got my API key plugged in, which I'll delete after the video. And the way that you can get an API key here is if you go to API keys at the top, then you're going to click add key. And then you're going to copy the API key that you get back. Like you can see, make sure you do copy it because the problem is once you've set this up, you can't go back and view the key, right? So if I click on that, I can't view it. I can only delete it. So just something to bear in mind there. But once you've set that up, we can then start using Flux to generate the images. So you can see here we're using q. fall. run and then the URL basically for creating images with flux on autopilot. So, we've got the URL for Flux. We've added the headers authorization. Plugged in the key there. And then for this next session, we're just going to click on send body. We're going to select the type as raw. Application/json is the content type. So, you select JSON there. All right. So, raw and then application/json. And then inside the body here, you're going to take this section here, plug it in. Boom. Ready to go. So, we're just going to test that step now. And as you can see, that's basically sent a bunch of requests over to Flux for each scene inside the prompts. Right. So, let me talk you through what we're doing here to generate this cheeky little video right here. We had to come up with the idea. The way that we did that was we selected this stage idea like let me show you what I mean. So we had to generate the idea which we did over here that's sent over to Google Sheets. Google Sheets created the prompts for the images and then you can see here we're generating the images in this section using a HTTP request and the models that we're using here are chat GPT and also Flux for now but we'll come on to more in a second. So you can see here we've selected the body. We're creating the images like so. We've plugged that information into our workflow and that's generated all these images. Now, if you want to see what has been created so far, let me show you how. So, if we go back to the AICS success lab and then you're going to click on blocks. So, this link right here. Then you're going to click on requests. And inside requests, you can see all these requests we've created for the Google bot. So these are basically like the images we're generating to start creating our short video, our faceless video. So if we go back inside here, we've sent the request over to Flux to generate the images. And now what we're going to do is we are going to add a new section which is just wait, right? So, it takes a minute or so, like you can see, for the images to generate. So, we're going to add a new module called wait. Just add a second there. Click test step. Make sure that works. You can see that's waited. So, we've got everything inside the queue. And then from here, we're going to click on get images. So, if we click on get images, which is another request, type that in. And then from here, we're going to copy the URL. plug that in. All right. So, this is the request for taking the images from Flux. So, what you can see here is we're doing the request. We're choosing the method as get, take the from Flux, add in the JSON format after that, and then we need to click on send headers. Inside the send header section here, we've got authorization, and we're just going to plug in our API key again. Okay. And then we also need to send body type. So let's click on that. And then from here we'll type in this into body section and just double check that. All right. So we've got the files ready to go. So what you can see here is these are the actual files generated by Flux that we've literally just generated a second ago, which is pretty cool. And so we can not just generate the images with a HTTP request, but we can also get those images directly from Flux. And now we're automating and we're getting pretty close to towards generating the videos and that sort of thing. Now another thing you can do here at the bottom is you can select options and then batching. And this will just help you avoid some errors. So, if you select options here and then from this section, you can just select 2,00 milliseconds, click test step, and that will just stop it from having any errors, right? So, if there's any errors or if you overload the server, etc., then you can bash URLs and it will just help you avoid any errors along the way. So, there we go. We've got the files back now. Boom. And you can see there's multiple pages here at the bottom and we've completed stage three. So just to recap here, we've created the idea, we've generated the video prompts, we've generated the images, and now it's time to start generating the videos. So what we can do

from this section is if we click on create video prompts, let me show you how. So we're going to duplicate this. We'll call this Stage four is create video prompts. Now we're going to click the plus sign. Just drag that down there. Type in chat GBT. Hit open AI. And then from this section, what we can do is we can click on analyze image. So what this is going to do, it's going to analyze the images that we've generated just a second ago. So you've got the images generated from the URLs, which you can see over here. We're going to drag that over. So we'll drag that over to image URL section. We're going to select the model as chat GBT4 latest. There we go. And then what's in the image? We're going to take that. Right. So this basically analyzes the images from our prompts and also the image URLs which we've got down here. So, I'm just going to grab the prompt from the previous part of the workflow and we'll cut that and plug it inside here. There we go. All right. So, again, this is a dynamic JSON prompt based on the prompts from before and the URLs from the images we generated. We'll just test that. And as an example, you can see for the GoPro monkey image that we generated the video, it's broken it down step by step. So it's like a young monkey grips a black GoPro with both hands, crouch low on a lichencoed rock, a soft sunlight brushes the tips of his orange foot. Right? So simple prompts, easy to generate, and we're automating it with this process. So if we go back to the example prompt and we compare that versus the output. So, a young monkey, you can see it's quite a small monkey right there, grabs a black GoPro, which we've got over here, crouched on a rock. There's the rock and the sunlight is gracing fur, which you've got over here. Right. So, it's basically mapping out exactly what we want inside the image. Now, it can take a while for that to generate, but you can see now we've got that generated. The reason it takes a little bit longer than usual is because you're analyzing these images which does take time. So from here we've created the video prompts and we're now going to do a HTTP request for video generation. And the model that we're going to use inside here is cling. So if we click on HTTP request then we're going to select post as the method like so. Then we're going to use this URL as the URL for generating the outputs. By the way, any part of this video, if you need the prompts, if you just want to steal them from me, I'm showing them on the screen. You can pause the video, take a screenshot, and just ask chat GPT to type them out if you need to, right? Or you can just download the template from school and then plug that in. And this school is free, right? So, inside the authentication section, we're good to go. And then inside the send header section here we are now going to select authorization. And again we'll take that value for the API key. Like you can see inside the send body section we're going to select raw application JSON and then we'll copy the body like so. So if we test that step now, and one thing you should do as you go along here is just make sure that you label everything, right? So for example, you see how we've got get images on stage three rows over here. Just going to do that now. Change this to get images. We'll hit rename like escape off that. And then if we go inside the get images section, we're now going to plug in the prompt that we found before. We'll keep going through. Xiao says hi. Shout out to you. And let's just wait for this to load. There we go. All right. Then we're just going to test this again. Just waiting for this to load. So bear with me. There we go. So now this should work. Let's just hit test step. Make sure this works. Okay. Yeah, there we go. Good to go on that. And I'm just going to rename this video creator so I know exactly what it does, etc. And then we're going to wait again. So I'm just going to add a wait section here. We'll wait a second. But again, if you were like generating the outputs here, it might take 4 minutes for cling to actually come up with outputs. So that's what we're doing step by step. And also in this section, you can see we're using cling video, right? Cling video. That's what we're using. Now I'm using the standard version. You can use the pro version which is even better to be honest. So if you want like better quality images or something like that like this was just a test run. The images still look good to be fair the videos. But if you want even better responses then you can switch your models right. So for example if we go inside the course again then we're going to go to requests. Inside the request section here you can see the inputs for the video and what we've created so far. But if we actually go back here, we can switch on the drop down to pro, right? Image to video pro. And that would just make sure that the images you're using are even better quality. So if you want the creme de la creme, but you're willing to pay more, then you would use video pro. If you want like the standard sort of usage, then you're going to switch to image to video. From here, we've done that. We've worked our magic. We're just going to wait a second here. Generate the outputs. There we go. We're good to go. And then what we're going to do is click on get video. Right? So we're going to add another HTTP request. Click get. Type in the URL to get requests which you can see right here. Now important thing to note here is if you have a space after the request URL, it's not going to work. All right? So make sure that you use the exact same format that we've got right here. And then we've attached a dynamic JSON request from the previous responses which is going to get the cling video step by step. Right now just to recap here, you've got get, you've got URL inside the settings as well. You would send headers using fields below. Then you can select the name and the value for your API key. Again, you ever you're typing stuff like this, especially if you're sending headers and stuff like that, you've got to get it precise. If you add the wrong keys and stuff like that, it's not going to work. So, for example, with the API key, we put key space and then we have the rest of the field right here. And then for sending the body, we'll select that. You can just duplicate it from the last one, but I just want to make sure it's super clear what we're doing here and how we're doing it. For sending the body, we're going to take the description from the prompt. And then we're going to put the batch in as one and 2,000 the interval time. Just test how it works. Some people are asking like how do you get access to the templates etc. So let me show you how this works step by step. So let's say you are watching the live stream. You can click on the video, find the description down here. Click on that and then inside here go to classroom AI course and then you can find the template and all the resources down here. So go to classroom and then AI course. Hope that makes sense. Just going to double check this, make sure we've got it all right. And also it does take a while for the outputs to generate on cling, right? So, if you want to see where you're up to in terms of generating the outputs from Clling, then we can go out to cling over here. We're going to go into requests. And you can see, for example, the videos ready to go over here. Pretty cool. Charles says, "Good to see your show. You seem to be doing good works. " Thank you very much, sir. And then I'm just going to change this as well to video. And there we go. Just one thing to bear in mind here when you're doing this, make sure you got everything set up correctly. Otherwise, it can break. It is temperamental. So, you want to make sure everything's matching nicely. Now, once you've done that, you'll see an example like this. So, you just test step at each stage. Here's an example. And then you're going to move on to the next stage. All right. So we get the video like so. Here's an example. You're good to go on that. Then you got stage five. So stage five is where you start

generating the sound. Now I'm using GPT4. 1 for this. Again, it is a little bit slow. So if you don't want to use it, you can just test something else. You can see an example of the output right here. And then this again is a chat GBT message model. We've got the prompt like basically it's just a prompt to generate the music. So for example, it looks at the image. It's a lively symphony of rainforest. Life unfills, leaves whisper overhead as a humid breeze stirs through the treetops. Beautiful and poetic. And it's got descriptions for each scene. Then from there, it's going to start creating the sound. Now this is another HTTPS request post. We're using audio for generating this. Plug in your API key again. and then send body select raw JSON and here's the example of the prompt that we're using. Then you're going to wait a second and then from there get the sounds. So what this does is it takes the audio from audio, you send the headers, you send the body, there's the descriptions, etc. and you're good to go on that. And that leaves us with stage six. Now stage six

you're basically stitching everything together. You're taking all the code then putting it into a format it can actually understand. So it's taking all the URLs downloaded it can download the video. So this is like a 5-second clip. It's taken all the 5 seconds that we've generated with cling and stitching them together. So you can see here what we've done. We've said code list elements run for all items language JavaScript and then here's the code to generate that. That's going to stitch the video using another HTTP request. So, post here's what we do the headers and then raw content type equals JSON. Again, you're just going to take the URLs like so. These are all the video URLs we're plugging in. So, you can just drag those in. Then you're going to wait a second and this gets the final video. So, for example, if we take a look at this output and we open this up, this is the final output of the video we generated. like you can see downloads it automatically. And then the final thing that we want to do is just plug that into our Google sheet. So here's an example of all the settings we're using, the idea, etc. Here's the code that we plugged in. And basically what that's going to do with the final output is it's going to match the column based on the idea. We're mapping it manually. Then it's going to look up that idea and plug it into our Google sheet. So if we go back to the Google sheet template, which you've got free access to, it will plug it directly into the sheet as you can see right here. So this will plug directly into the Google sheet and have the URL. So that's basically it for automating reels. The next part that

I'll come on to later is how to actually automate the posting of all of this directly to social media. So just to recap here, we've come up with the idea, the video prompts, generated the images, generated the videos, created the sound, stitched it all together, then plugged that URL into a Google sheet, and with the Google sheet, we can actually take the outputs and then automate posting that directly to social media as well. So here's an example, right? So if we look at this one, you can see that for the idea of a monkey, we've got the caption, production, the environment, and then the final output of the URL, which takes us to this video that we've just downloaded. And you can automate posting that directly to social media, like you can see, which I'll come on to in the next video. So that's basically it. That's how to automate and create reels with just one idea. You don't even need to come up with the idea yourself and it's super powerful using this whole spreadsheet. So, thanks so much for watching. If you want to get free access to all the templates, workflows, etc., you can get that inside the AI SEO success lab in the comments in the description. And inside this community, we post a bunch of like notes from the courses and the videos that we do. So, you can see like client and stuff like that. And then if you want to upgrade and get for example like coaching weekly calls etc feel free to get that inside the AI profit boardroom link in the comments description. So you just click on that and then that'll take you to the AI profit boardroom and you can sign up and get coaching an awesome community of people focused on really scaling with AI and also all of our best courses agents and workflows and that sort of thing. So feel free to get that link in the comments description. If you want us to build this out for you or any sort of SEO automations as well, feel free to book in a free SEO strategy session link in the comments description and we can help you out with that too. So thanks so much for watching. Appreciate it and I'll see you on the next one.