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.

Sharbel A. · @sharbelxyz
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 Sharbel A.'s most watched videos.
Most replayed moment at 13:32
3.9x that video's typical replay level
built-in memory, which is great. Here is how I would think about memory providers. Memo is interesting if you want a dedicated memory layer for personalized AI agents. It focuses on extracting, storing, linking, and retrieving memories efficiently. Their
Said at 13:25
Most replayed moment at 6:37
5.8x that video's typical replay level
trading strategy. Tests it. If it's better, it keeps it. If it's worse, it discards it and tries again. So, that's exactly what I built. Okay, here's the system we have at play. I gave it two years of crypto data,
Said at 6:31
Most replayed moment at 1:53
3.7x that video's typical replay level
let's install it together. Okay, so for step one, we need to actually start by installing Bullpen's CLI so that we can actually do everything that we want to do. And for that, let's first open Claude and put in the dangerously skip
Said at 1:47
The graph counts replays. It does not show where viewers stopped watching.
Words
1,913
Runtime
12:39
Speaking pace
151wpm
Reading time
8min
151 words per minute, below the 160 25th percentile of 349 measured videos. That distribution comes from the 349-video hook study.
Opening (first 30 seconds)
I gained 2,000 subscribers in the last 4 weeks without writing a single script [music] myself. I got YouTube monetized, made $14 on day one and $19 on day two, but here's what actually matters. I built an AI agent that does the content strategy work for me. This is Nova and I'm showing you exactly how it works so you can build it for your niche. Hi, I'm Charbel. I'm a founder at Unfungible and
76 words, the words spoken in the first 30 seconds at 151 words per minute.
Free, no signup. See how the first 30 seconds hold attention, with rewrites.
Sentence shape
| Measure | This transcript |
|---|---|
| Sentences | 143 |
| Average words per sentence | 13.4 |
| Longest sentence | 54 words |
| Questions asked | 9 |
| Sentences containing a number | 26 |
Most used terms
Filler phrases
26 in total: actually 10 · like 8 · basically 3 · uh 2 · I mean 1 · literally 1 · you know 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.
Free, no account. See 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. Or run it on the words above first.
Free · No login · See a sample audit first if you prefer.
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.
I gained 2,000 subscribers in the last 4 weeks without writing a single script [music] myself. I got YouTube monetized, made $14 on day one and $19 on day two, but here's what actually matters. I built an AI agent that does the content strategy work for me. This is Nova and I'm showing you exactly how it works so you can build it for your niche. Hi, I'm Charbel. I'm a founder at Unfungible and a creator trying to hit 10,000 YouTube subscribers this year. 4 weeks ago, I had 800.
Now, I'm sitting at 3,000, but here's the problem. I don't have time to be a full-time YouTuber. I mean, I run a company. I'm building other things, so instead of hiring someone or spending 6 hours a week on content strategy, I built an agent to do it for me and I'm going to show you the exact file you can paste into your own OpenClaw so you can build the same agent for whatever you're building. So, before Nova, my content strategy mostly consisted of guessing.
I'd upload a video, hope it would perform, and sometimes it did, but sometimes it didn't. I'd look at the analytics afterwards and think, "Okay, that one hit 600 views instead of 300. Why?" But by then, I was already working on the next video. There was no system in place, no learning, just me throwing ideas at YouTube and hoping something stuck. The prep time was also killing me. It took me 4 to 6 hours per video just to figure out what to film, write the script, organize my thoughts.
That's a full day of work added on top of running my company. And I tried the typical content advice, post consistently. Okay. I did that. Focus on one topic. Sure. But nobody told me how to pick the right topics, how to know if [snorts] an idea would actually resonate with my audience, how to write a title that would actually convert clicks. I needed something that could do three things. Number one, look at what's actually working in my space for other creators.
Number two, analyze what's working in my own channel's history. And number three, tell me what to film next. Not have to guess. That's when I realized I could build this myself. So, here's what Nova does. And I want to be specific because the mechanism is the whole point of this. First, Nova monitors my entire niche. She's looking at creators and OpenClaw and AI agents, AI creators, agency and clients. Basically, for creators in my niche.
And then, she tracks which of their videos outperform. Not all of their videos, but what it calls the outliers, the ones that hit a 2x, 3x, 5x their average view count. For example, if you look at this creator right here or this video that it's highlighted, this creator typically gets 900 views per video, yet on this video that he posted 14 days ago, he has almost 200,000 views. That's a more than a 200x his average view count.
Why does this matter? Because those outliers show what the algorithm actually rewards. If someone builds an audience in say crypto or tech or AI and one video gets 10,000 views while their others get 500, something in that video hit. The hook, the title, the mechanism, and Nova is studying it. She's looking for all the outliers that are in my niche. Oh, would you look at that? I am I am an outlier in my own niche. Second, Nova reads all of my transcripts, every single one of my videos.
She's learning how I talk, where I put humor, how I structure ideas, my tone, and my language patterns. She's reverse engineering my voice. Then, she compares what she learned about my voice to what actually performed on my channel. Which videos did I make that got traction? Oh, look at this one. Got 10,000 views on YouTube and almost 200 views on Twitter. She's seeing what videos flopped. She's connecting the dots between my voice patterns and my actual results.
Third, this is where it gets very useful. Before I film anything, Nova asked me a set of questions. You can see this interaction that I had with her yesterday. I want to film how to build a YouTube growth agent with OpenClaw tomorrow. Is this the best video for me to film? She looks it up. She comes with a result and then she asks me a set of questions. She interviews me. She says, "What does the agent actually do in the demo?
Who is this for, creators or builders? Do you want a watch me build it live format or I already built it, here's how it works and any results you can show?" And you can see me over here giving her a 7-minute voice note with all the answers. Basically, like an interview. And the lovely thing is I can literally do this when I'm anywhere. I can do this when I'm at the gym, when I'm out grocery shopping, when I'm busy, I can just send her a voice note replying to all of her questions and then she can use that data and information to compile the script for me.
The biggest problem I found with watching creators that use AI is a lot of the time you can sense that the AI is just, you know, just putting blanket statements out there and you can feel that it was AI-written. The nice thing about this setup and what I personally love about it is that it's not coming up with the script from scratch. It's not hallucinating just facts and uh things to say on the video. Rather, it's looking to me for information in order to use that information in the video itself.
Everything, for example, in this video are things I've outlined to it in this voice note right here. And from my answers, she generates video ideas based on the actual data, not guesses. You can see her from this very voice note, she goes like, "First, you got monetized?" That's the actual video, not how to build a YouTube agent. It's how an AI agent got me YouTube monetized. That's the hook, my guy. Nobody can resist that title.
That's not something I would have thought of. I would have just posted a "How to build YouTube agent" and the video would have probably flopped. But instead, she gave me title suggestions and pivots that are essential. Typically, she gives me between two to five title options for each idea, not one title, ranked by SEO potential and click-through probability based on what she learned from the niche analysis that she did.
Then, I pick one, for example, this one uh is the one I picked for this video that you're watching, and she generates the script. We iterate for like 2 minutes and then I have a title that's research-backed instead of just a gut feeling. From there, she writes a full script that is in my voice. For example, this video right here. Hook. I gained 2,000 subscribers in the last 4 weeks without writing a single script myself.
The hook for this very video right here that came started from a from her researching what is out there to me choosing something I like, to her interviewing me, to me responding to the interview, to her taking it, extracting that information from the interview, and laying it out in that script. And the scripts she generates, don't get me wrong, they're not perfect. I still edit them, but it's 80% of the way there. That used to take me an hour and hour and a half of just staring at a blank page.
Now, it's like 10 minutes of cleanup for me. And here's the thing that makes this different from just using ChatGPT. This agent is specific. It's not writing generic AI tutorials. It's writing tutorials that fit my channel, that sound like me, that are informed by what actually converts views on my channel. Okay, I'm going to show you this and I want you to guess. Just guess when Nova started helping me out. Just take a wild guess. 4 weeks in, I've gained over 2,000 new subscribers.
The videos I make with Nova now average 1,500 to 2,000 views each. This not viral by YouTube standards, but for a channel that was getting 40, 90, 250, 600 on a good day before Nova, that's a 4 to 10x improvement. And YouTube finally monetized me yesterday. It was my first day generating revenue on YouTube. I made $14.5 yesterday, $19 today, and it's just going to keep going. It took me almost a year to get to 1,000 subscribers before I built Nova.
I got to 3,000 in 4 weeks after. And yes, I know the current generated revenue is peanuts, but here's the point. I wasn't trying to monetize YouTube. I built an agent, I used it, I got results, and now YouTube is paying me. But, real talk, the YouTube money isn't the play yet. The real number is this. I'm making a thousand to a thousand five hundred per month from X creator payouts from my content. That's happening because I'm posting consistently.
My voice is clear, and I know what content actually works because Nova is telling me. That creator revenue alone covers all my AI agent costs and then some. I'm basically getting YouTube content for free while I build an audience on X, which is where the actual money is happening for me right now. So, the ROI is real. It's not theoretical, it's real cash flow backing up the system. Now, here's where you come in. I built Nova for myself, but I realized I could open source the entire thing.
So, I put it on GitHub. It's free, and the link is right in the description of this video. You don't have to be a YouTuber. You could be a writer, a podcaster, a builder shipping product, anyone with a voice and an audience that you want to grow strategically instead of by accident. The repo has the instruction file, a setup guide, and a walk-through on how to install it in your own open claw. You just clone it, follow the read me document, and you have the same agent running for your niche in under 10 minutes.
No code required. The file tells the agent what niche you're in, what your voice sounds like, who your audience is, where to monitor for trends, and how to generate ideas specific to your work. You just fill in your details, and it's yours. The link is in the description, so go ahead and get it. Oh, and one more thing, while I'm spending my time building bots that iterate automatically, the reason I can do that is because I'm not manually having to post on X and YouTube every single day.
That's being handled much like you just saw. So, if you're a founder and you want to actually build your presence online without spending 3 hours a day writing tweets or making videos, that's what we do at Unfungible. It's called the timeline takeover. Go check it out at founderfunnel.com, and subscribe for more agent builds and real creator data. I'll see you in the next one.
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.