Big Idea

The argument in one line.

A single structured prompt, or the same prompt saved as a Claude Code skill file, can drive an entire AI production chain — fact-checked research, script, visuals, voiceover, and assembly — into a finished Vox-style explainer video without a manual editing pass.

Who This Is For

Read if. Skip if.

READ IF YOU ARE…

You run or want to start a faceless YouTube channel and need a repeatable way to turn a topic into a finished explainer video without hand animation.
You already use Claude Code and want a concrete example of chaining an MCP tool through a multi-step creative pipeline.
You want a documentary-style visual format for business, news, or educational content but don't have an animator or editor.

SKIP IF…

You want a fully free workflow — Higgsfield runs on paid credits once any free trial ends, and longer videos cost more credits.
You need hands-on control over individual shots — this pipeline hands scene composition to Claude and Higgsfield rather than a hand-built storyboard.

TL;DR

The full version, fast.

This is a step-by-step screen recording of building Vox-style paper-collage explainer videos entirely inside Claude Code. The creator connects Claude to Higgsfield through an MCP server, giving it access to image generation, video generation, AI voice (including voice cloning), and music from one account. He pastes a long structured prompt — built on a six-block, one-object script formula — that tells Claude to research the topic, cross-check facts against two sources, generate the visuals, and assemble a finished video, then shows a shortcut: a packaged skill file that runs the same pipeline from a one-line request. He closes by comparing format switching and per-video credit costs.

Free for members

Chat with this breakdown — free.

Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.

Create a free account →

Chapters

Where the time goes.

00:00 – 01:15

01 · Cold open: the payoff

Cuts between three finished Vox-style explainers (Strait of Hormuz oil war, AI officiating the World Cup final, a cheap Chinese model rattling markets) before revealing they were all built from one prompt.

01:15 – 01:51

02 · Tools you need

States the two-part setup: Claude Code (desktop app or browser) plus an MCP connector that can generate images, video, voiceover, and music.

01:51 – 03:48

03 · Connect Higgsfield MCP

Creates a Higgsfield account, copies the MCP connector URL, adds it as a custom connector inside Claude Code, and authenticates in the browser.

03:48 – 05:18

04 · Free prompt pack + the six-block structure

Points to a free Skool community for the prompt pack and reveals the framework behind every Vox explainer: six roughly-ten-second narration blocks built around one recurring physical object.

05:18 – 09:11

05 · Running the 60-second prompt

Pastes the full prompt into Claude Code with a real topic, watching the model research the story and cross-check facts against two sources before generating images, voice, and assembling the final video.

09:11 – 12:08

06 · Skill file shortcut

Introduces a packaged 'Vox Motion Graphics' skill file — attach it and ask for a one-minute explainer with no other context, and Claude reads the skill and runs the whole pipeline itself.

12:08 – 14:00

07 · Format tweaks and costs

Interrupts an in-progress generation to switch from a vertical 9:16 preset to a custom 16:9 landscape layout, and flags that longer runs consume more Higgsfield credits.

14:00 – 15:54

08 · Voices and cloning

Shows the default AI voice roster, including cloned voices already saved under the creator's own name, and explains how to record a short sample to clone your own voice.

15:54 – 19:47

09 · Results and wrap-up

Plays both finished renders in full — the Strait of Hormuz explainer and an AI-power-demand explainer — then closes noting music was left out on purpose, with more resources in the free community.

Atomic Insights

Lines worth screenshotting.

A single long-form prompt can generate a complete Vox-style explainer video — script, paper-collage visuals, animated charts, voiceover, and transitions — without touching an editing timeline.
The underlying structure of a Vox-style explainer is six narration blocks of about ten seconds each, built around one recurring physical object that carries the story.
Instructing the model to search the web, cross-check every number against two independent sources, and default to conservative figures when data is disputed produces a script grounded in verifiable facts instead of hallucinated claims.
A packaged skill file lets Claude Code generate the same style of video from a one-line request, instead of requiring the full step-by-step prompt every time.
Switching aspect ratio from 9:16 to 16:9 mid-generation forces the model to regenerate a custom visual style key rather than reuse the vertical preset, which changes both output and cost.
Voice generation defaults to a stock AI voice, but the same pipeline supports uploading a short recording to clone the creator's own voice for narration.
Background music was deliberately left out of some renders because AI-generated music is a common complaint among viewers, not because the tool can't produce it.
Longer videos consume proportionally more generation credits, so render length is a budget decision rather than a free toggle once a free trial runs out.
The entire pipeline runs through one connector: Claude Code's MCP integration with Higgsfield handles image generation, video generation, voice synthesis, and music inside a single account.
The video's own cold open is made of outputs from the exact one-prompt process it goes on to teach, using the finished product as proof before explaining the method.

Takeaway

One prompt, one skill file, one finished explainer.

WHAT TO LEARN

A long, well-sequenced prompt — or the same prompt saved as a skill file — can carry an entire AI production chain end to end, but only because a fixed six-block, one-object script structure keeps every generated piece pointed at the same story.

01Cold open: the payoff

The three finished clips shown before any explanation are proof-of-concept for the method — the video does exactly what it teaches, which is the strongest form of demonstration for a workflow claim.
Leading with output before process front-loads the payoff, so viewers commit to the tutorial length before being asked to set anything up.

03Connect Higgsfield MCP

Connecting an external creative-generation service to Claude Code as an MCP server turns Claude into an orchestrator that can call image, video, voice, and music generation without leaving the coding chat interface.
Authentication happens in the browser, not the terminal — Claude Code hands off to a normal consent screen, the same pattern most MCP connectors use.

04Free prompt pack + the six-block structure

The reusable structural insight, independent of any tool, is that a documentary-style explainer is six roughly-ten-second narration blocks wrapped around one recurring physical object that carries the story visually from open to close.
That fixed structure is what keeps a fully automated script coherent — without it, letting an AI both write and illustrate a topic tends to drift.

05Running the 60-second prompt

The prompt explicitly sequences the model's work: research and fact-check first, generate narration and visuals second, assemble last — ordering the steps prevents the model from illustrating a claim before verifying it.
Grounding a script in a real, personal source produced more specific narration than starting from a blank topic in the demo.

06Skill file shortcut

A skill file collapses a long, step-by-step prompt into a standing instruction Claude reads on demand — the user only has to state the topic and length, not the whole method every time.
Because the skill file already encodes which MCP tool to call, the model doesn't need to be told to use Higgsfield — it infers that from its own saved instructions.

07Format tweaks and costs

Changing aspect ratio mid-generation isn't just a resize — the model regenerates a custom visual style key for the new format instead of stretching the vertical preset, so format decisions are cheaper to make before generation starts than after.
Longer runtimes cost proportionally more generation credits, so render length is a real budget decision once a free trial runs out, not a free toggle.

08Voices and cloning

Voice cloning only needs a short sample recording and a few minutes inside the tool's audio settings — a one-time setup cost that then applies to every future video in that voice.
Named custom voices sit in the same picker as stock AI voices, so switching between an anonymous AI narrator and a personal voice is a one-click choice per project.

09Results and wrap-up

Both finished renders skip AI-generated background music entirely — a deliberate choice made because AI music is a common source of viewer complaints, not a limitation of the tool.
The visible quality bar — synced narration, hard cuts every few seconds, a consistent paper-collage look — is achievable without an editing timeline, but only because the script structure and voice were locked before generation started.

Glossary

Terms worth knowing.

MCP (Model Context Protocol): A protocol that lets an AI coding assistant like Claude Code call out to external tools and services — here, Higgsfield's image, video, and voice generation.
Higgsfield: An AI media generation platform connected to Claude Code via MCP to generate images, video clips, voiceover, and music for the explainer.
Skill file: A saved instruction file that Claude Code reads to reuse an entire multi-step workflow from a single short request, instead of retyping the full prompt each time.
Vox-style explainer: A documentary-style video format popularized by Vox's YouTube channel — paper-collage visuals, one recurring physical object as a visual metaphor, and tight narrated blocks of about ten seconds each.
Voice cloning: Uploading a short recording of your own voice so the AI voice model narrates future videos in your voice instead of a stock AI voice.

Resources

Things they pointed at.

01:51toolHiggsfield MCP ↗

04:05linkAI Workshop Lite (free Skool community — prompt pack + skill file) ↗

15:06productJARVIS AI Assistant (paid Skool community) ↗

14:17toolElevenLabs (voice cloning model inside Higgsfield)

Quotables

Lines you could clip.

00:05

“I created these Vox style AI animation videos with one prompt. It turned into a war zone.”

cold-open hook line that promises a dramatic payoff before revealing the method→ TikTok hook↗ Tweet quote

07:34

“Search the web for the real facts on the story. Cross check every number against at least two independent sources and list your sources at the end. If the number is disputed, use a conservative figure.”

a concrete, reusable prompting instruction that forces the model to fact-check itself before writing a script→ newsletter pull-quote↗ Tweet quote

12:52

“Switching to 16 by nine, that means generating a custom landscape style key instead of a vertical preset.”

shows the model reasoning about a format change mid-generation instead of just following a fixed template→ IG reel cold open↗ Tweet quote

The Script

Word for word.

Read-along

Don't just watch it. Burn it in.

See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.

metaphor

I created these Vox style AI animation videos with one prompt. It turned into a war zone.

In June, Washington and Tehran signed a deal to stop the fighting. And three weeks later, drones started slamming into cargo ships. Oil punched back past $100 a barrel, and American jets hit about 140

targets in a single week. The World Cup final is finally here, and there is one player on that pitch nobody is talking about, artificial intelligence.

This year, it touched absolutely everything. Every offside call runs through 16 tracking cameras and a smart ball that feels the kick 500 times a second. The referees just hear a beep.

So what does it look like when a $100,000,000,000 company panics? It looks like this.

Wednesday, a Chinese lab nobody was watching ships a Frontier model at a third of the price. So in this video, I'm gonna show you step by step how to create these and I'm also gonna give you all the resources completely for free including this prompt pack and this skill file that you can upload to your Claude or Claude code to generate different formats of this Vox style AI videos including long form, short form, 16 by nine, nine by 16 videos with your own voice narration, videos with AI voice narration, and everything else you need.

Alright. I don't wanna waste too much of your time. Let's jump right in.

Alright. So now to create these, we need a few different tools and a few different process setup before we jump jump in into one prompt generation. First of all, I'm gonna be I'm gonna be using this obviously inside Claude code.

Uh, I'm gonna use the desktop app so feel free to install that. That's the best case best way to use this but you can use this inside your Claude browser as well. So once you have your Claude set up and ready to go, the next thing is we need to give Claude the ability to generate images and videos.

Right? And also voice over narrations and music as well. Right?

Because we want to have just Claude do everything and we just sit back and tell it what to do. So in order to do that, we're going to give Claude an MCP for Hicksfield.

Now, Hicksfield is kind of like a one stop shop for, you know, image and video models and then also our audio as well where, again, like I said, we're if we want to do our own voice over or if we wanna just use an AI voice and also the music as well if you wanna generate videos with music. So it's kind of like a one stop shop for all of it.

So in order to give plot code access to Hicksfield, we need to connect this to the Hicksfield MCP. So go ahead and click on the description. It's gonna bring you right here.

Sign up for an account. I think they have a free trial, so take advantage of that if that's still there. Once you sign up for an account, it's gonna land you right here.

Hopefully, if you click on the link in the description. If not, you can just click on the MCPCLI. It's gonna bring you the same exact spot.

Once So you're here, all you have to do is just make sure you're on Claude and MCP and you just copy this link right here. Okay? So once you do that, you're gonna go back to your Claude code, you're gonna go to the plus button here in the bottom, you're gonna click on connectors and you're gonna go to manage connectors, you're gonna click on add, add custom connector and here you can do Hicks field or name it something and then just basically paste that URL that we just copied.

You're gonna click on add, mine's not gonna add it because it says a servo with this URL already exists but yours is going to add this and it's gonna bring you to this page right here. Here, yours obviously, mine's already connected.

Yours is gonna say connect. Once you click on connect, this is gonna open a new page, uh, on your Chrome and on your browser and it's gonna ask you to authenticate and allow ClotCode to use the Hixfield MCP.

You're gonna click on that and that basically gives you authentication and when you come back here in your cloud code, click on the plus button here, go to connectors, you should have now Hixfield MCP installed right here and toggled on.

So that's basically it as far as the tools that we're going to use. Right? We're going to use cloud code and the Hixfield MCP to generate all of this.

So now the next step the next, uh, step is the generation part. Now, of course, like I mentioned in the beginning of the video, you guys will have access to this, uh, prompt pack. Right?

So this Vox style explainer prompt pack. This is gonna give you all of the different ways you can generate these Vox style AI animation videos for different niches. So I have a bunch of different examples here.

You can go through and read it yourself when you get a chance. In order to access this, I'm gonna put the link in the description again. It's a free community.

Please click on that link that says free community. It's gonna bring you right here. It's called AI workshop lite.

So it's gonna bring you an about page. Just click on join. It's free again.

Once you're here in the community, once you join. Right? You're gonna go to the classroom section.

You're gonna go to YouTube resources and you'll have access to the prompt pack right here, this prompt packed and then also in a skill file that I'm gonna talk about in a little bit. All of that is going to be inside the YouTube resources folder. Okay?

So very, simple. Obviously, if you get stuck, let me know in the comments below, put a common saying prompt or something and I can send you a direct link as well. But it's pretty easy once you join, you'll be able to see all of the resources there in the classroom section.

Okay. So once you do that now, like I mentioned whoops. Let me open this up quickly.

So yeah. So afterwards, like I said, open this because I give you kind of instructions before you start what are the different sources that are in this prompt pack and what can you generate. Right?

So this just gives you a really good template to start with for different topics and niches. Right? Everything is kind of or everything is kinda separated into six blocks for one object.

Again, I don't wanna go through the prop back. You guys can take a look at it yourself. So let me go ahead and actually just copy one and we can go ahead and, uh, create one of these so that way you can take a look at what that looks like.

So I'm gonna copy this first one, one story, one finished video out. Right? So let me go ahead and make this a little bigger so that way you can see the whole thing.

So let me go ahead and paste this. So this one in particular is about let's say you already have a topic in mind. Right?

Like if you were, uh, doing something about a particular business or some kind of a new cycle or whatever it is. So this just basic basically builds the entire sixty second box style motion graphics explainer video about that particular topic and that's why right here it says in bold like big letters, it says your topic.

So this is where you would actually put your own topic. So for instance, this is Jarvis, my AI assistant, my second brain. So everything I do inside my computer, everything kinda gets saved here.

So if I want some kind of a topic that I wanna work on, it can create a animation style, this Vox style animation. I could just basically do that.

So let me go ahead and do that. Jarvis, can you pull up all of the information related to the Jarvis project I've been working on? So now Jarvis is gonna pull up all of that information so that way I can just get something up all the information, sir.

You can read it yourself or if you'd like, say read it out loud and I shall. Thanks, Jarvis. Alright.

So this is basically everything that I've been working on this week. Obviously, yours is gonna be look different. So let's say I wanna grab some information about this Jarvis v five word word roadmap that I've been working on.

By the way, for those of you who are not familiar, so Jarvis is my AI assistant. I've been improving his ability to do more and more. So this week that that's what I've been working on.

So I can basically just take this. Right? So I can copy all of this and just paste it inside my clot code here, uh, in my topic and it's gonna go ahead and generate it.

Or let's say you wanna create something about, I don't know, about a news topic or something like that. Right? So I'm just gonna do, let's say, about the Iran war because obviously that's kind of like the news cycle right now.

So all I have to do is just now press enter. So now this is gonna go ahead and follow all of the different, uh, tasks here at each different step based on this big prompt and it's gonna go ahead and assemble a sixty second video for me, uh, and at the end, all I have to do is just basically download that and upload it to my social media.

So as you can see the first step right now is browsing the web to find out the recent news. So as you can see right here, the first thing it's going to do is step one research.

Search the web for the real facts on the story. Cross check every number against at least two independent sources and list your sources at the end. If the number is disputed, use a conservative figure.

So I'm actually giving Claude the first step is actually research the internet, make sure about this particular topic you find the relevant information and then you won't move on to the next step. And then afterwards, like I said, the next steps are kind of going and creating the narration so this time obviously I'm not giving it my own voice.

I just kind of left it to its own determination to use some kind of a AI voice that's inside of our Hicks field. So it's gonna just create a, uh, voice over itself but I'm gonna show you how you can create your own one as well.

By the way, I'm using Opus five. Opus five is the new model that you just released but obviously this works with any, uh, model. I haven't to be honest, I haven't tried Sonnet five yet but I've tried with Fable five and Opus five and it works really really well.

So but feel free to try it with other models as well. Okay. There you go.

So now it's done with the research part. Now as you can see, it's using the Higgs field MCP.

So it says widget from Higgs field generate image. So it's gonna use Nano Banana Pro and it's gonna go ahead and generate all the images and resources it needs including the voices. Right?

Because it's just gonna generate all of that itself and then at the end, it's gonna combine everything together and give us a sixty a sixty second Vox style animation video in that particular topic. Alright.

So while this is generating, let's go ahead and move on to another, uh, another format so that way you can see what other formats you can create with this. Okay.

So now let's say you want to create like a two minute or three minute or five minute video. Right? A long form video instead of like a sixty second short form content.

Now for that, uh, let's say you don't have a topic in mind or you just want to create something relevant about today's news or like I said, even if you have a topic in mind, you don't want to have this long prompt where it just goes ahead and kind of describes each step by step. So for that, for those of you guys who who don't wanna do that, I'm giving you access to this voxmotiongraphics.

Skill file. So, this is basically just a skill file that you can give clot code where it will take care of everything for you. So, all you have to do so I'm just gonna go ahead and bring this over here So you can download this and bring this over here.

Again, downloading is from the free community there. You'll have access to the same skill file.

So all you have to do is just attach it and now you can say something like create a one minute explainer video about a relevant topic related to AI. Right? Or even even if you don't even have to, uh, say related to AI or whatever topic you want.

Right? You can just enter this and it will figure out everything itself. So let's say let's say related to AI, I just entered this.

Now you'll see that it's gonna go ahead and read the skills. So it says I'll build a Vox style motion graph explainer. Let me first read the skill contents properly.

So it's gonna read this skill dot m d file, right, because now I'm giving it a prompt that has no context. I'm not giving it just before like before step by step explanation of what what it needs to do or what it's supposed to do or should it use the hex field m c p or not.

Now this is kind of like the shortcut. It's gonna go ahead and actually grab all of those steps from the skill file so that way all you have to do is just give it a topic or you don't even have to give it a topic.

You just say create a one minute, two minute, three minute explainer video about something. Right?

Like and it'll just go ahead and do some research and ask you details about, hey, is this is there something specific you want me to focus on or could be anything that's hot in the world of news today or the world of AI today or whatever it may be. So now it's gonna go ahead and go through that process and all the details that's inside the skill file and generate for you.

So as you can see right here, it's loading voices now so it looks like it figured it out, it uses Hicks field. Right? I never explained to use Hicks field because in that skill file, all of the instructions are there and automatically knows to generate the assets, the voice, the music, the images it needs to use the Hicksfield MCP.

So the first thing it needs to do and it will check to make sure that you have giving you have given clot code access to the Higgs field MCP and if it and if you have, it's gonna go ahead and generate that. Now you can see in the bottom right here, so it says plan topic locked going straight in generation topic. Why?

AI is showing up on your electricity bill. Right? Angle, data centers added 23,000,000,000 to US power bills.

Format, six blocks, ten seconds, sixty seconds or ten seconds each, sixty seconds 19 by six vertical. Right? So now it's gonna use the automatically.

It's gonna use 19 by nine by 16 which is a short form sixty second. But if you want to have it generate something different.

Right? So let's say we can stop this and say, hey, generate a 16 by nine one and a half minute, you can certainly do that. So let's go ahead and try that.

So let's go ahead and add the skill again and this time I'm gonna say and make sure it's in 16 by nine format.

Okay. So let's see if it catches that. So like I said, the point is you just give it a few instructions and it's gonna go ahead and create all of that for you without you really getting involved in kind of the small details of, uh, the asset generation, the voice over, and everything else.

So it says switching to 16 by nine, that means generating a custom landscape style key instead of a vertical press preset. Right? The three earlier submissions were intercepted by a preset per hour.

So now it's gonna go ahead and actually do that for you. Uh, so now there you go. So it's rendering the image, it's generating the image and it's gonna go ahead and do everything else in these.

By the way, like I said, I only did one minute. You could do whatever. You could do two minute, three minute, whatever it is.

Of course, it's just gonna cost you more money. It's gonna cost you more Hicksfield tokens. So make sure that you are paying attention to that and depending on what plan you're on on Hicksfield, make sure you're, you know, not wasting your credits on something that you need.

So go ahead and take a look at the pricing for Hicksfield if you you want to take a look at the different options there. Alright.

So let's give these a couple of minutes. By the way, as you can see, it's already generating the voice so it already chose this Sterling as the voice over.

Right? So if you wanna change this but we can click on this so you can see what it sounds like. Electricity in America

is up roughly 40%.

There you go. Right? So it's kind of an AI voice.

Now Hicksfield has multiple options when it comes to voice selection. So if you don't like the voices, you can actually unclog itself. You can tell them, hey, change the voice to something different.

So if you go to Hicksfield, if you're a Hicksfield account, go to audio here. Uh, you can actually even add your own voice. Let's say you want your own voice narrated here and it's very simple to do.

You go to audio, you can select the model. I've used 11 labs v three before but you can use any of these other models. I'm gonna click on 11 labs.

So you have multiple options here. So you can do voice over, you can do a change voice. Change voice just basically means that you uh, when you upload a file that already has a voice, let's say a video, you can change it to something else.

Voice over right here, can clone. So I've already cloned my voice but you can click on this and click on a create a custom voice and you'll be able to drop a clip of yours or you can just start recording, it's like a two minute recording and tell you exactly what to read and you have your voice call and it's very very simple.

So then when you come back and finish the video or let's say even in the beginning, you can just say, hey, use whatever name you put for your voice. So mine's called zubert dot m p three, I think.

So I just said I would just tell them use my own voice. Right? So then it will generate that video animation in your own voice over, uh, so that way it sounds like a bit more personal.

But again, a lot of people, you know, they don't care. They just have they just want to have an AI voice so you just you can just let it, uh, choose the voice by default. That works too.

Okay. It's gonna take a couple of more minutes for this to finish. Now like I mentioned before, the classroom section in the free community is where you'll have access to the skill file and then also the prompts.

And if you want to actually create your own Jarvis AI assistant because this is one of the most fun project I've ever worked on. Right now, we were in, like, version five. That's gonna be inside the paid community because I've spent a lot of time and money on generating these so that way you can just install Jarvis as your own personal assistant.

Alright. So the videos got done. So as you can see, this was the Iran war war one.

So let's go ahead and open this. One fifth of the world's oil squeezes through a channel just 21 miles wide

and this month, it turned into a war zone. In June, Washington and Tehran signed a deal to stop the fighting, and three weeks later, drones started slamming into cargo ships.

56 ships have been attacked this year. Tanker traffic has collapsed to nearly zero, and Iran has declared the strait closed.

Oil punched back past $100 a barrel, and American jets hit about 140 targets in a single week.

But here is the real game. Iran is not trying to sink ships. It wants to own the road and charge the toll.

And whoever blinks first, the bill lands in the same place. Every tank of gas on Earth now carries a piece of this war.

Alright. Very cool. So as you can see, it does a really good job of putting everything together, kind of the zoom in effect going through different images and videos and assets, the voice over, everything is kind of in sync together.

Now this would have taken a good few days if you were to build this manually using After Effects. So you can see how incredible this, uh, method of creating these types of videos are.

And again, like I said, that was just like a one minute video. You can basically create any amount. Of course, like I said, it's gonna cost you more money but you can imagine how powerful this is as far as the storytelling aspect of it.

So that was one. This one is done too. So this was the one that we've said, we gave it the skill.

Right? That skill dot m d file. It said create a one minute explainer video about the relevant topic related to AI.

So I went ahead and generate it. So let's go ahead and open this now. Okay.

So let's play this. Data centers just added $23,000,000,000

to American power bills and almost none of that computing was doing anything for you.

Electricity in America is up roughly 40% since 2021, and the fastest growing new customer on the grid is artificial intelligence.

Data centers use about 6% of America's power today, and that is forecast to reach 20% by 2035.

That is one unit of energy in five from every coal plant, gas plant, and solar farm feeding machines instead of homes.

The strange part is that for years, those same data centers actually pushed rates down by spreading grid costs across more customers. The build out outgrew that math. So the cheapest power now belongs to the machines, and the bill quietly lands on your kitchen table.

Alright. Very cool. So you can see this one's also, uh, really well done with that AI voice over, uh, of the Sterling voice, right, the AI voice.

Now, of course, this one these these ones are without music in the background. So if you want to include music, you can certainly do that. I did it on purpose because a lot of people don't like the AI music that it generates.

But to be honest, you know, some some of the ones that I've created with so for for example, this one Thirty eight light years This one has music. A planet See?

Just exhaled.

And for the first time in history, we watched another world breathe.

It is called LHS eleven forty b into space. Yep. So it has sound effects, voice over, and a music in the background.

So again, there's lots of things you can do here. So, anyways, I don't wanna make this video too long. So, hopefully, you guys found that helpful.

Let me know if you guys have any questions in the comments below. Make sure you like and subscribe because I've got a lot more videos coming up in this space that you wanna miss. Thanks for watching, and I'll see you on the next one.

The Hook

The bait, then the rug-pull.

The video opens by playing three finished explainers back to back — an oil-strait war, AI officiating a World Cup final, a Chinese lab undercutting a $100B company — before admitting all three came out of one prompt, then promising the exact prompt pack and skill file for free.

Frameworks

Named ideas worth stealing.

05:04concept

Six Blocks, One Object

Every Vox-style explainer breaks into six narration blocks of about ten seconds each, built around one recurring physical object (an oil barrel, a shipping crate, a server rack) that carries the story visually from open to close.

Steal forany documentary-style explainer script, even outside AI video tools — it's a scripting structure, not just a prompt trick

CTA Breakdown

How they asked for the click.

VERBAL ASK

04:05link

“Please click on that link that says free community... you'll have access to the prompt pack right here, this prompt pack and then also a skill file.”

Soft, resource-adjacent CTA delivered mid-tutorial rather than at the end — the free prompt pack doubles as the lead magnet for the creator's paid JARVIS community, which only gets a brief mention later near the voice-cloning segment.

MENTIONED ON CAMERA

01:51toolHiggsfield MCP ↗

04:05linkAI Workshop Lite (free Skool community — prompt pack + skill file) ↗

15:06productJARVIS AI Assistant (paid Skool community) ↗

FROM THE DESCRIPTION

PRIMARY CTAWhere the creator wants you to go next.

Storyboard