Getting the transcript
Reading the captions from YouTube. A video nobody has opened here before takes 10 to 30 seconds; this page fills in on its own.
Getting the transcript
Reading the captions from YouTube. A video nobody has opened here before takes 10 to 30 seconds; this page fills in on its own.

Maciej Dziuba · @MaciejDziuba
This video has no Most replayed graph yet: YouTube shows one only once a video has enough views. These are the moments viewers replayed most in Maciej Dziuba's most watched videos.
Most replayed moment at 2:04
8.8x that video's typical replay level
links are going to be down below. So, then just open up the terminal or the command prompt if you're on Windows, and then CD into whatever folder you want to work out of. Then you can just open up Cloud. So, I'm going to do Cloud with dangerously skip permissions. Then we're going to start the ComfyUI server. Again,
Said at 1:58
The graph counts replays. It does not show where viewers stopped watching.
Words
5,551
Runtime
22:36
Speaking pace
246wpm
Reading time
23min
246 words per minute, above the 201 75th percentile of 349 measured videos. That distribution comes from the 349-video hook study.
Opening (first 30 seconds)
Most editors, creators, and business owners are going to miss this shift entirely. They think that prompting ChatGPT means they're using AI, but the real leverage is in using agents, which will edit an entire video for you while you're out for your walk and have it ready for you by the time you're back at the office. What if I told you that we're going to do just that in this video? We're going to get Hermes agent to trim our video end to end. It will create animations using Hyperframes, just like the one you're seeing on screen right now, and it can even spin up Blender renders and touch design workflows just from a single sentence. And look, I'm a
123 words, the words spoken in the first 30 seconds at 246 words per minute.
Free, no signup. See how the first 30 seconds hold attention, with rewrites.
Sentence shape
| Measure | This transcript |
|---|---|
| Sentences | 334 |
| Average words per sentence | 16.6 |
| Longest sentence | 101 words |
| Questions asked | 10 |
| Sentences containing a number | 19 |
Most used terms
Filler phrases
168 in total: actually 43 · like 41 · basically 23 · I mean 21 · you know 15 · kind of 11 · literally 6 · uh 3 · right? 2 · um 2 · sort of 1.
A literal whole-word count of the same phrase list the Prepublish browser extension uses, so a phrase inside another word is not counted and a phrase used in its ordinary sense still is. It is a count and not a judgement.
Run the check on the words above: where attention is likely to drop, with a rewrite for each weak line. The free check shows the scores and the one issue costing the most.
What this transcript is
Every word below is the caption track YouTube publishes for this video, pulled from the video itself and reproduced unchanged. It is not Prepublish's writing, not a summary, and not a re-transcription: it is the video's own published captions. English captions, generated automatically by YouTube, in the video’s original language. Source: the video on YouTube. A channel that would rather this page did not exist can ask for its removal through the contact page, and it is removed.
Most editors, creators, and business owners are going to miss this shift entirely. They think that prompting ChatGPT means they're using AI, but the real leverage is in using agents, which will edit an entire video for you while you're out for your walk and have it ready for you by the time you're back at the office. What if I told you that we're going to do just that in this video? We're going to get Hermes agent to trim our video end to end.
It will create animations using Hyperframes, just like the one you're seeing on screen right now, and it can even spin up Blender renders and touch design workflows just from a single sentence. And look, I'm a video editor with over 600 million views on my portfolio, and even I'm making sure to stay on the cutting edge of this stuff. Are these tools better than a senior video editor? No. But the direction that this is heading in is undeniable, and I'd rather be early than late.
Now, you're probably thinking, what if I'm not technical? Look, here's the thing. All you need to do is be able to speak in a natural language. You don't need to be some AI expert, and you don't even need to be an editor. Now, what's cool about the agent that we're going to set up today is that you can actually be walking down the road, think of a video idea, quickly record it, upload it to Google Drive, and have Hermes edit absolutely everything while you're still on your walk.
Trimming, animations, captions, and even uploading. And if the first pass isn't perfect, you can simply send it some revisions in voice notes and have it apply that feedback. But the real reason why you should actually set up Hermes right now is because it gets more powerful the more you use it. Because Hermes is actually a self-improving agent. So again, even if you've never set up a terminal in your life or you've never even edited a video, just block out 20 minutes because I'm going to show you how to set everything up so that you can actually use Hermes as your video editor today.
And look, I wasn't kidding Hermes is really easy to install because it's literally just this command. All you got to do is just copy it. There'll be a link for this below. Open up your terminal and paste it in. Now, I'm going to go with the quick setup, and I'm going to choose OpenAI Codex because I already have a ChatGPT subscription, which means that this will basically be free. So let's choose OpenAI Codex, and we're going to say yes.
Default model, of course, GPT 5.5. I'm going to run it locally. Then it's asking for the working directory, so I'm just going I'm going a folder in here and press enter. Enable sudo support, yes. Password, and we are going to set up messaging, no. So, it's asking to launch it. I'm going to say yes. Okay, I'm just going to say hi just to test and see if Hermes is working, and as you can see, we have Hermes running locally on my MacBook.
So, the first thing I'm going to do is I'm going to set up Telegram because, as I said, I want to be able to walk down the road, record a quick video, and send it into Hermes. And if Hermes is running here locally on my MacBook, how am I meant to do that, right? So, I'm just going to type in plain English, let's set up Telegram, help me step-by-step. And as you can see, we have super easy And so, step one, open Telegram.
Step two, search for BotFather. So, here we see BotFather. Let's open that up. So, we should go {slash} newbot, send that. Choose a name for your bot. I'm going to say Hermes Agent. Now, it needs a username. I'm just going to say hermes_agent_mache. Okay, it just needs to end in bot, so {underscore}bot at end. Then, it's going to send you an API. Just copy that and send it back into Hermes. Then, type in {at} userinfobot, and copy the ID it gives you.
And then, just paste your user ID into Hermes as well. So, we have Hermes now on Telegram, meaning you can just message it from your phone. Great. So, next up, we're going to actually connect Google Drive to Hermes Agent so that we can send larger files as videos and have those edited. So, I'm going to send off rclone config into Hermes. Hello. It's asking which cloud or storage backend we're setting up, and we're setting up Google Drive.
So, obviously, just send that off. Then, it's going to just pop open a window for you. You don't even need to open this manually, and it's going to be a Google Auth sign-in. So, just sign in to the account that you wanted to connect Google Drive for. Once you do that, you're going to see this success screen. So, just go back to Hermes. We're going to say it said success. By the way, if the pop-up didn't appear for you, you can always just click on this link and then authorize to Google Drive.
And there we go, Google Drive is configured successfully in rclone. If you want to test it, you can just send off rclone list remotes, and as you can see, we have Google Drive connected. So, now let's just create an output folder for Hermes so that when I finish these videos, it's going to upload all the videos inside of a specific folder. Again, I'll leave a link to everything in the description below, so you can just copy and paste these commands, or you can just see them on my right now.
Now, as you can see, we have the Hermes folder in my Google Drive now, and we can have this output folder here as well. So, if you actually want Hermes to use a specific file in your Google Drive, then hit that. You can simply right-click, share, copy the link, and paste it in and say use this file, trim it down, add effects, add captions, whatever. It's basically already done. But now, let me move on to actual use cases and show you some cool examples that you can use Hermes for video editing.
So, the first effect I want to show you is how to create these cool Tree Blue One Brown style animations directly from Hermes. They're seen on screen right now. I mean, if you ask me, these are pretty sick, and honestly, I don't even know if I could do some of these animations myself, and I've been editing for quite a lot of years now. So, now Hermes Agent basically has these things called skills, which are like plugins essentially, that you can download and import into Hermes Agent to give it more powers.
And so, in order to create those animations that you just seen, we're going to use the Manim video skill. And installing skills is very easy, because all you need to do is simply tell Hermes what to install. So, we're going to say install Manim and its system dependencies, FFmpeg, MacTeX. Tell me when it's done, and restart yourself. If you want, you can even copy the link from the GitHub and paste it in for extra certainty.
There's no way to write skill, but I trust it, so I'm going to just send it off. And as you can see, we have FFmpeg and Manim installed. There was one thing that wasn't installed, so I just had to add the password manually, and I added that. And by the end, we have basically everything installed. So, now it's time to send off our first prompt and actually get a video edit back. So, we're going to use Manim to animate the Pythagorean theorem written out, then a right triangle drawing itself with squares on each side that morph into the equation.
So, I'm not sure how good this is going to be, but let's just send it off to Hermes and see what it cooks up. Okay, sometimes if there's approvals needed, meaning, you know, you need to actually take over and let Hermes do something, you're going to get something like this. So, I'm just going to say always allow, because obviously, these agents have a lot of access. They have a lot of powers, so you do sometimes need to be careful, especially if you're running them directly on your MacBook like I am.
But this is why I recommend you actually use very powerful models. Oh, we have an animation. Let's pause this for a second. This is why I recommend you use the best models because you don't want to risk your AI agent deleting important files, moving something, leaking some some data. There is risks involved, so do be careful with these models. But by using these best models, you are essentially reducing that risk for yourself.
And of course, you're going to get better results, which let's see how this came out. Nice. Nice. Nice. Okay, this is really clean. I mean, I'm just Yeah, that's low. I mean, I'm just considering like how long this would actually take me to create in After Effects. Obviously, there's little tweaks that can be made here and there. But I mean, yeah, this is pretty good. >> [snorts] >> The shapes morph into the text. It's basically exactly what I asked for.
And again, these visuals are going to improve. You are basically the 0.1% of editors right now, not even top 1%. Most editors aren't using these tools because they just look at the, you know, bad aspects, things that need to be improved, the things that it can't do yet. But by being on the cutting edge, you're setting yourself up and your future up to actually be able to edit 10 times faster, 10 times better, and land 10 times more clients.
And I'm not even overexaggerating. I mean, this is where the future is going. And keep in mind, this is just the start. I'm about to show you so many more use cases to use Hermes agent for. But keep in mind, this is just one of many use cases that AI is actually good at editing videos. I mean, as you can see here, I use it for re-motion, comfy UI, hyperframes, creating videos like Kurzgesagt, TouchDesigner, Blender, viral education reels, and many more use cases that people don't even realize you can create with AI nowadays because they remember AI from last year, which is basically like working with a different tool a decade ago.
So, if you are looking to land clients, become a faster editor, and actually stay on the cutting edge of all these AI tools, then I do suggest you join us in the editing evolution, where you can find a bunch of different materials, all the prompts I use, all the setups I have for every single AI agent that actually helps with video editing. And I reveal how exactly I actually built such a portfolio with over 600 million views, working with over 55 different YouTubers, 19 of them actually being verified, meaning they had to sign in on their YouTube channel in order to verify they worked with me.
I mean, you can vet me if you want. You can look through my profile. I am a legit video editor, and I do have the proof to back it. So, if you're enjoying content just like this one, I'm sure this will be an investment worth making. Okay, also, I just wanted to upload this to Google Drive. Obviously, you have the preview here, but all you got to do is tell it because remember, we have Google Drive connected. So, I just said upload this to Google Drive.
And as you can see, we have our output in Hermes output in our Google Drive. For the next use case, I want to actually trim a video because a lot of times you might need to, you know, record a video, and there'll be a lot of pauses, there'll be a lot of retakes, etc. So, what if we can just send some rough footage directly into Hermes and get it basically trimmed for us? Well, let's find out. So, again, you can just copy and paste this prompt, which basically has a bunch of different dependencies that help our agent edit code.
By the way, if you're wondering how long the previous output took, it only took 3 minutes, which was kind of surprising. I mean, the setup took longer than this. So, it's not actually as slow as you might think. Like, here, it took 1 minute to basically install everything, so that's cool. So, I have a video already on Google Drive that I wanted to trim. So, I'm going to paste in the link, and then I'm just going to say this prompt.
Take this video, download it with Reclone, then trim it cleanly. Remove all silent pauses that are any sentences or phrases that the speaker repeats, giving it more context, like where they restart a thought or say the same thing twice. Transcribe the audio with timestamps, so you can actually identify the repetitions, and make the cuts smooth so that the audio flows naturally without obvious jumps. And then, instead of sending a second prompt to upload it, I just included that in this prompt so that it automatically uploads our video.
So, for context, this is the video before. Clearly, there's a lot of dead air, me getting ready. Remotion is hands down the most powerful AI native video editing tool right now. >> [clears throat] >> Okay, as you can see, clearing my throat, pause, looking over at the script, etc. Like, there's a lot a lot of dead air, right? So, this is going to be a video that I'm going to use to get it trimmed. Okay, so we got a response, uploaded to Google Drive with a link, and it cut down 90% of the video.
So, keep in mind, even if this video isn't perfect, it saved me about 90% of my editing work if I am to trim it myself. And also, keep in mind that I basically didn't have to do anything. I just simply sent her the link to the video and told her to edit. And also, as you can see, self-improvement review was just created, meaning this is Hermes basically improving itself over time. This is where the more you use it, the better it will actually get.
It'll understand your preferences and how to actually edit the way you want it to edit. So, if we check out the video, we can see it's 14 seconds long, so let's play a bit of it. Remotion is the most powerful video editing tool right now. But the problem is that 90% of people are using it wrong, or they just don't know its full potential. Yeah, I mean, it basically cut out all my editing notes and a lot of different pauses.
So, again, this saved a lot of time. But again, this self-improvement is actually OP because even if this doesn't come out good the first time, the more revisions you give it, it'll understand your preferences. I'm not over exaggerating. This is why you shouldn't give up on the first try, and this is why it's important to be on the cutting edge, so you understand this kind of stuff. Give us some feedback, give us some reviews, and then watch it actually implement that feedback next time around.
Now, if you're wondering what's better than one Hermes agent, well, multi-Hermes agents, or multiple Hermes agents, I guess. This is what we're going to cover next. How to actually use the Kanban view inside of Hermes agents so that you can keep progress of multiple agents running at once, even if they're focused on the same goal. So, what we're going to do is just paste this command into the terminal on Mac, and then confirm.
Then simply paste in Hermes dashboard, and we can see Hermes agent opened up this nice desktop app. So, on the left over here, we have Kanban. This is basically where multiple agents will live in like this nice user interface. So, you can see you know, what's being worked on, what's in the to-do, what's ready, what's in progress, etc. Okay, so for the next use case, I decided to just basically do something completely from scratch, not even using my own footage, and I decided to go with a 30-second Instagram reel about how I use AI to create branded content for clients, and then just give it some extra prompt engineering.
And then down here, as you can see, we have the link. It's obviously uploaded to the Google Drive, and they even added an English voiceover and some electronic music and some animations. So, let's check it out. I use AI like a creative production team. I turn a client's brand into a visual system, colors, tone, blocks, and format. >> Stop. To be honest, the video in the background is actually kind of cool. I like this little animation, but I wouldn't use this for motion graphics.
The reason being is because we have things like Hyperframes and Remotion. Speaking of which, this is what Hyperframes actually is. It is a way to essentially create motion graphics and edit videos using code instead of pixels. And as you can see, some of the outputs here are pretty insane. And what we can do is connect Hermes agent to Hyperframes so that it creates these motion graphics and edits videos for us, which is what we're going to do right now.
Now, if you're feeling overwhelmed with all of these new AI tools, then I suggest you subscribe to the channel because 93% of you guys are not even subscribed, and it does help out the channel a lot more than you think, and it's a completely free way to show your support to the channel. Well, it's helping out far more than you think. Also, it's going to curate your FYP page so that you actually have more high-quality AI videos just like this one on your FYP page.
So, if that's something you're interested in, then be sure to subscribe. And if you did, thank you very much. Now, to install Hyperframes, what we got to do is say Hermes skills install official creative Hyperframes, and it's going to begin installing the skill. Just like that, it took literally a couple of seconds to install. I forgot why I'm not using Telegram, but we can use Telegram, okay? Going back to here. So, we have the skill installed.
So, now I'm going to paste in my prompt. It seems like a pretty long prompt, but keep in mind I do type these in plain English, and then I just tell AI to optimize it for AI agents. So, again, I'm utilizing multiple different AI tools to help me accomplish my goal here. Okay, as you can see, we have a generation done here. So, let's see. It did render a 60 FPS Hyperframes video. Let's check it out. This is much better than what we just seen, by the way.
Okay, there's some small tweaks, but I'll show you in a second how to actually make granular adjustments if you want. But this is already much better, and this is where you need to understand what tool to use for what purpose. If you just seen the previous use case where I showed you how to create some sort of motion graphic just using Hermes, you would have been like, "Okay, Hermes is pretty shit." But now, with that you can connect Hyperframes to it, shows you that there is much more to it.
I mean, look at this. This is sick. Even the gradient of the blue down here, very nice. Yeah, I mean, this is also, you know, one-shot prompt, very straightforward. If you did watch my Remotion or Hyperframes video, you would have seen that this tool has a lot of potential, okay? And Hermes Agent has a multi- multiple different skills to optimize some of these. The only issue that I find with basically every platform that I use it is too granular control.
For example, right here, the text is obviously covering the other text. So, this is so simple. It's literally just moving it down a little bit, but I don't want to have to re-prompt it. It's way too much effort for such a simple use case. That's why I'm building my own app to solve this issue, because it is an actual issue I'm facing day-to-day when I'm trying to generate these um motion graphics. As you can see here, this is literally just one prompt away or like one little movement away from adjusting something.
And that's exactly what I can do here. I can kind of like move something even just a little bit around or, you know, add a keyframe to it, move it down, add another keyframe to it, and now it's like moving left and right, whatever. So, changing positions, changing colors, opacities, completely deleting stuff, changing text, and it's just very easy to do. Like, if I want to move the text here, have it changed, and I say change it to blue, I can do just that.
I can change everything around. This is exactly why I'm building this app. Also, once this finished generating, I can obviously type in a prompt and actually enhance it, because I basically wanted to make this as seamless as possible for people who don't understand all these technical details about hyperframes and Remotion, which is also here. Yeah, basically you can generate both Remotion and Hyperframes at once. And by typing a prompt, you can actually enhance it, and we're going to use the best skills and the best optimization techniques for creating a prompt for Hyperframes or Remotion.
If you're interested, I'll leave a link to this down below in the description. But, you do need your own Cloud Code or Code X to actually run locally, because I am fixing the Cloud SDK right now. Okay, now what I'm going to do is I'm going to paste in the link to the trimmed video, so you can watch my prompt right here. That's all I'm going to send. I'm just going to send it the link. I think he has access to my Google Drive, so it should be good.
Okay, while this is playing around, I was just kind of testing on my app and this is why I really do believe that you should use Hyper Frames and Re Motion because even though most people are saying, you know, Hyper Frames is way better, Hyper Frames is can do this, can do that. Check this out. Re Motion actually made a much better, in my opinion, motion graphic on the left over here. It just looks much more fluid, you know.
A quick tip, even if you're not generating these motion graphics with my app, you should still add assets because this is what made this animation feel so much more high quality is just the fact that I added an asset. Same goes for obviously brand colors, fonts, etc. So, obviously I make it really easy by adding this add asset button. But, you can literally just drag and drop an asset into the prompt box and tell Hyper Frames to use your specific asset for a specific thing.
Okay, let's check this out. Re Motion is hands down the most powerful AI native video engine. Okay, I mean, I'm not sure what happened here, but there is zero motion graphics. I'm definitely not happy with this result. I guess it over optimized on the TV static old visual and under optimized on motion graphics, which were far more important. Potentially with my prompt, I mean, I did I was very vague with the motion graphics.
I just said generate full screen motion graphics. So, okay. Okay, so what I just done is I basically pasted it. I mean, you can read it here if you want. Basically, me just telling what exactly is wrong because look, I'm not going to lie to you when something is not working out correctly. Like, this is Like, so let's see if I can make it better, you know. So, I'm going to send off this longer prompt, which Perplexity cooked up rather than just my plain English prompt cuz this is a little bit of a harder use case.
I'm going to interrupt it though just so that it actually uses this video, okay. But, um these use cases are a bit harder with these motion graphics and stuff. So, maybe it is better to optimize it. But, again, I I mean, I don't want to be promoting my app like this whole video, but this is exactly what I'm trying to add here is this enhance button so that you can write a prompt in plain English like an apple falling from a tree and when you click on the enhance feature here, it should optimize your prompt to actually use Re Motion best practices, to use the correct Hyper Frames skills, and to give you the options to actually choose uh between the best kind of presets and assets that Hyperframes and Remotion has to offer.
Because as you've seen here, it is kind of just a waste of time sending off a prompt because you're going to get this one as a result. So, these optimized prompts are definitely the way to go. Okay, so as we can see, we do have a lot more motion graphics, a lot more text. A bit overdone, but it is what I asked for. And as you can see, we have context on what exactly Remotion even is. Captions for the whole video. The CRT or CRTV, whatever it is called, effect on the actual video.
This is pretty cool as well. This one's actually really good. Like this is something I would uh download just as an effect in and of itself to import into actual videos. Just a little color scheme here and the pixels are pretty nice. And yeah, I think there's too much text, not enough visuals. That's one problem these AI tools have a lot is that they over utilize text and they just don't use enough visuals. And there's small tweaks like the background here, which again, I should probably just use my app for this so I could scale this really easily.
But yeah, overall, this is much better than before uh because of the prompt. Okay, since you're making it this far into the video, I assume that you are already familiar with skills, how to install them. I mean, there's a bunch of different skills that you can install. So, just use your imagination for, you know, what kind of skills to add and what you can actually create with them. But I want to kind of show you how to go a little bit above and beyond.
This, for example, is a very cool skill that we could just install. It's very easy to do. But as you can see, there's not that many models. This is File AI and it is basically a website that allows you to generate different images, different videos, etc. And as you can see over here, we have way more options for way more different models that we can use compared to the skill on Hermes. So, even though the skill will be easier, the MCP, meaning kind of like a link connecting File AI to a Hermes agent, will be a lot more powerful.
So, if there are skills that don't have, for example, all of the models like we see here or they just have less features than the actual app itself, I recommend you try to set up an MCP, which is what I'll show you next. So, firstly, we need to get an API key. So, in file, you can just all go over here to API keys and then add a key. So, once you get your key, just copy it and then simply say add export file {underscore} key equals and then paste in your key and then just add this little bottom prompt down here as well.
Okay, as you can see, everything is installed and we have to obviously {slash} restart so that Hermes restarts and picks up on the key. And so, I'm just going to say use the file AI API key to generate an image with X Grok Imagine of a video editor getting replaced by AI cuz as you can see here, we have obviously X AI is Grok Imagine inside of File AI, but we don't see it here. If I type in Grok, you can see that it's just not available.
It's not a supported model inside of the skill. So, this is actually a use case that is valid because as I said, it's just not here. And as you can see, we have an image generated with Grok Imagine even though the Hermes skill didn't allow it because we used an API key. Now, I want you to actually understand what we're doing here because this isn't just a tutorial where, you know, monkey see, monkey do. I'm trying to explain to you like my thought process behind these things and how to utilize these tools.
Like connecting an API key is a very powerful method to get access to essentially any app. Like we can use Replicate instead of File AI. We can connect ComfyUI to create workflows. It's also a similar process for connecting Blender MCP so you can use Hermes to create Blender compositions for you. I mean, the use cases are really endless. I mean, let me know if you do want a use cases video for video editors using Hermes, but you get my point.
I mean, this is kind of what I'm doing on the channel. It's showing you what's actually possible with these tools right now and what you should be using these tools for so that you stay ahead of the game. Like previously, learning something like Blender was extremely time-consuming and required a lot of trial and error, required a lot of tutorials to create some, you know, good quality animations, and of course, it required a lot of time to manually tweak everything.
And I'm not saying there's no space for something like Blender, but I'm saying that everyone understands AI is replacing video editors because let's be honest, everyone understands that AI is replacing all human jobs. So, if you actually do believe that, why are you being arrogant and thinking it's not going to replace editors? And I don't mean within 20 years time, I mean within the next year or two. Because this space is evolving faster than any other space in my opinion.
Like, let me show you an example from literally 10 months ago, okay? This is 10 months ago, and this is where the AI motion graphics space was. Obviously, I was covering it. At the time, it was cutting-edge technology. But, we can see this is pretty basic stuff. It's just three boxes, three icons, and some text. It is extremely basic. And keep in mind, this was basically cutting-edge stuff. I'm not overexaggerating. I mean, my whole channel is about covering video editing AI tools, and this is the best I could find.
Now, we have motion graphics like this. You know, this is a terminal nice transition shot. You know, logos bouncing, animations. If you know how to make this more powerful, by the way, you can use your own videos, your own images, your own icons, logos, color schemes, etc. And these are custom images. Also, these agents can go on different websites, take screenshots of things, actually get them factually correct, and import them into your projects.
Like, they're getting really powerful. And my point is that this was only 10 months ago. Remotion basically turned everything around because we're not using pixels for editing videos anymore, we're using code for a lot of it. And code is extremely powerful, which a lot of people don't realize can be used for video editing. But, as I mentioned, you can also use it for things like creating Blender animations, which is exactly what I done in this video.
So, if you do want to see how I use AI agents to actually create Blender projects, just click on the video, which is going to be somewhere up here, and I'll see you there.
The words are the caption track's own and nothing is reworded or re-transcribed. Paragraph breaks are placed between sentences so the text reads as prose.
Free tools for your own script: paste a draft and see where it stands before you record it.
Paste your draft and see where viewers are likely to drop off, with a rewrite for each weak line.
Paste the first 30 seconds of your own draft for a hook score and rewrites.
Check your draft against YouTube's advertiser-friendly guidelines before you record it.
Read this channel's public videos and transcripts, and download a writing brief for it.