you2idea@video:~$ watch sduaTkhIm_w [47:44]
// transcript — 750 segments
0:01 Hey, welcome to the definitive resource all about skills. I currently run a
0:05 business that does over $4 million a year in profit and I manage it primarily
0:09 through AI agents and skills. And I teach over 2,000 people how to do the
0:13 same thing. I think a lot of demos and walkthroughs of skills right now are
0:17 like really flashy and they're typically centered around the personal assistant
0:20 angle, but a lot of people are leaving tons of money on the table because
0:23 they're not applying them to specific business use cases that actually tend to
0:27 produce large returns on investment. And so what I wanted to do in this resource
0:30 is I just wanted to walk you guys through what that actually looks like.
0:33 I'm going to start by showing you guys examples of a bunch of skills that I
0:37 currently use in my $4 million a year business. Skills that other people are
0:40 currently using across a variety of different industries like ecom service
0:44 businesses, consulting, and so on. And then I'm going to show you guys how to
0:46 build them. And we're not just going to build cookie cutter skills that are
0:50 glorified personal assistants, but skills that you guys could actually
0:52 implement in your business to do the work of dozens of people in just a few
0:55 minutes. Okay? So everything you need is in the link in the description. Uh it'll
0:59 give you guys these skills so you guys could use them across your own business.
1:02 It'll also give you guys the framework that you could use to to build new ones
1:05 regardless of whatever your use cases. No fluff. Let's get right into it. So if
1:09 you don't know what I'm looking at right now, this is anti-gravity running five
1:13 different cloud code plugins. Um if you guys are new or unfamiliar with this
1:17 sort of interface, just check the link that I'm pasting right over here. It'll
1:21 walk you through everything from like what the icons here mean to, you know,
1:25 how to communicate with Claude code and Gemini and and other, you know, Agenta
1:29 coding platforms literally from the ground up. So, you you'll know
1:31 everything after that. But assuming you guys are somewhat comfortable, um, I
1:34 want to show you guys five different skills today and then I'll throw in a
1:37 couple of additional bonuses. The first one over here is going to follow up with
1:40 all of my leads. So, if you guys are running a business at really any level
1:44 of revenue over a few thousand a month, you probably have something akin to like
1:48 a list of leads or a list of contacts. This type of thing is typically referred
1:51 to as a customer relationship manager. And in general, what you do is you store
1:55 leads at different areas along the pipeline from when you first met them
1:59 all the way to close deals and money is in your pocket. And so I have an example
2:02 one here with a bunch of leads at different stages. Meeting booked, Jimmy
2:06 Clint, proposal sent, Sarah Chen, Jessica Park, Aisha Muhammad, and so on
2:10 and so on and so forth. And basically, right now, in order for me to move this
2:13 pipeline forward every day, I just have to manually check in with all the leads.
2:17 Well, I built a skill that effectively automates the entire process. In order
2:21 to trigger this, all I need to go and do is go back/f follow-up nurture. Now,
2:26 what this skill does is it goes through my pipeline. It identifies it captures
2:30 all of the conversations I've ever had with any of these prospects. So, it
2:32 actually goes through all of my email chains and so on and so forth. And then
2:36 it uses a couple of templates to personalize check-ins um depending on
2:40 where they are. So, you know, if a meeting has been booked but we haven't
2:42 actually attended it, it might check in with Jimmy and say, "Hey, Jimmy, how's
2:45 it going? just wanted to check in in our meeting in a couple days. Really excited
2:49 to have it. You know, if it's to Sarah and maybe we sent a $2,500 quote for a
2:53 specific type of product, I'd be like, "Hey Sarah, you know, hope you've had a
2:57 lovely week so far. Just wanted to circle back on X, Y, and Z proposal. Let
3:00 me know if you have any questions." The whole key here is we do this really
3:03 informally and then really casually so they think it's us. I think a big
3:06 problem that a lot of these automated follow-up services typically run into is
3:10 they just use really like non-human language. It's very clear and you know I
3:14 think prospects can tell when an LLM does the following up for them. But I
3:17 also like the fact that this just does it all in one shot. And so the workflow
3:20 is literally like you just wake up in the morning where your salesperson wakes
3:24 up in the morning um they go/f follow-up nurture. They immediately clear out
3:27 their whole pipeline of follow-ups and then they can just focus on acquiring
3:29 new business. What's also cool is it does all this in the tone of voice and
3:34 in the same chain of conversation as the initial email. And so in this case, this
3:38 is an email that I sent to Mr. James that said, "Hey, James, circling back on
3:40 the proposal, let me know if I can answer any cues." This is another one
3:44 that I sent to Priya that said, "Hi, Priya. Hope you had a great week
3:46 checking in on that test brief. Let me know where we're at." You know, this
3:49 just continuously customizes based off where the person is. And then it also
3:53 gets to pull in, you know, context and uh more or less everything that you
3:55 currently talked about. So, you're not just repeating the same thing. This also
3:58 goes through and then it lists all of the follow-ups that I've made to every
4:01 single person. So, you can see here we're signing off slightly differently
4:04 between them. We're using reply chains and stuff like that to stay within the
4:08 same thread. Um, and I'm going to give you guys that for free. I use it
4:11 basically every day. This second flow is basically a oneshot thumbnail generator.
4:15 Um, I have this funny image of John Ham or Don Draper and Madmen if you guys are
4:19 familiar. And this image went viral a little while ago. It's all about vibes
4:22 and whatnot. And so I figured, you know, as somebody making videos on vibe coding
4:26 and aentic platforms, I would try and reproduce the thumbnail. But instead of
4:29 doing it manually, I'm actually just going to say reproduce the John Hem
4:34 thumbnail. Use similar lighting etc. And then what it's going to do is it's going
4:37 to take this thumbnail or this image which has already been created. Then
4:40 it's going to go get my face and then it's just going to superimpose my face
4:44 onto that image in a really realistic looking way. It's also instead of just
4:47 generating one variant going to generate multiple so that we can pick and choose
4:51 which one looks the best. And also because not all AI outputs are perfect.
4:54 This is an example of the finished product and we did that just in a few
4:58 seconds. Um I also ran it again so you guys could see even more. I mean, just
5:02 like going back and forth between this pretty freaking close, right? However,
5:05 there are some important differences. The people in the background are
5:08 different. The lighting is just a tiny bit different. Uh, and so on and so
5:11 forth. And I mean, we've we've generated more. I have this running again in the
5:14 background. This one's with like a darker lighting and different tone. Uh,
5:17 I don't know. This one here is sort of like matrix style blue and sepia. And
5:21 so, all these are just a little bit different ones that it's not just like a
5:24 one to one rip of the source image. Um, but also because, you know, I get to
5:27 pick and choose variants that I like the most. And so, in my case, I really like
5:29 this one. And I think this like did the most justice to my face. And this is
5:32 probably the one I'm going to use for my video. By the way, despite the fact that
5:35 I hate when people do this to me, I'm going to do it to you. If you like this
5:38 sort of thing, please subscribe. Uh something like 69% of you guys aren't
5:41 subscribed for some reason. So, that's two out of every three people that I'm
5:45 talking to right now. And it just significantly impacts my channel's
5:48 ability to grow. So, if I've given you anything, any value at any point in time
5:51 over the course of this video or any others, just click that button for me.
5:55 You'd be doing me a big solid. Okay, back to the video. This next one here is
5:58 all about scraping leads. And so I could actually just say, "Scrape me 50
6:03 management consultants in uh I don't know, let's just say Arizona." And what
6:07 this will do is it'll go on LinkedIn Sales Navigator, which is currently the
6:10 highest quality place that you can get information about uh real human beings.
6:15 It then creates a bunch of LinkedIn search URLs. And in case you guys didn't
6:19 know, LinkedIn Sales Navigator allows you to filter people really granularly
6:23 based off things like what their job title is, um where they live, you know,
6:28 their various seniority levels, I know their industry, and then basically I
6:32 just get a list, not within Sales Navigator, but in a Google sheet with
6:37 these people and then all of their email addresses. And like pre-existing
6:41 solutions exist to do this for Apollo and Appify, but I don't think there's
6:44 like currently a big one for LinkedIn Sales Navigator. The reason why that's
6:47 valuable is because there is no better source right now for B2B lead data than
6:51 LinkedIn Sales Navigator. It's the most current. Uh most other services just
6:54 scrape from there anyway. And as a result, I get to go straight to the
6:58 source and then just ask a system for what I want in natural language. Uh so
7:01 within, you know, 3 or 4 minutes, I end up with a nicely optimized list of
7:05 high-quality leads with probably like the highest deliverability out of more
7:08 or less any source that you could get today. And then you end up with a really
7:12 highquality list of people literal with their full names, first names, last
7:17 names, email addresses, and so on and so forth that we've sourced basically
7:21 directly from LinkedIn Sales Navigator. Um, this is done pretty effectively via
7:25 cost as well. And then the fact that it's just going to occur in the
7:28 background while I do other tasks, it can retest its leads. It can figure out
7:32 like the best filters to use and so on and so forth. Um, eliminates a fair
7:35 amount of my own day-to-day work as somebody that runs like a growth
7:38 management agency. This next one here writes my cold email campaigns. So, for
7:41 those of you guys that don't know, a cold email is something that goes to
7:45 somebody who you haven't actually talked to before, hence why it's cold. And then
7:49 it tries to sell them on some service. And so, I run cold emails just as my
7:53 day-to-day for a couple of clients. Essentially, what I commonly have to do
7:56 is I have to iterate and change cold email campaigns based off of a new
8:00 client's information. And so, what I want to show you guys is basically how
8:03 you can automate this process. I now going to use the backslash convention.
8:07 I'll go cold-empaigns. Sorry, I think that's actually a forward
8:11 slash. And this will basically ask you, hey, what are the details of the client?
8:14 And now I'm just going to voice transcribe and say, the client is 1
8:19 second copy. Uh their offer is they want to write free 500word sample articles
8:23 with no strings attached. Find the similar campaigns and then use those as
8:27 inspiration. I'm then going to feed that in. And then my skill is going to go through and then
8:33 essentially duplicate high-erforming pre-existing campaigns within my cold
8:38 email platform. Uh and then rewrite them so that they're as similar and high
8:41 quality as humanly possible for that given niche. Then sets it all up in my
8:45 cold email platform. So again, a step that previously might have taken me 30
8:49 to 60 minutes just logistically is now taken care of. And then obviously all
8:52 the offer management and stuff like that is also handled entirely autonomously.
8:55 So yeah, I mean this takes something that maybe 3 or 4 hours into something
8:58 that takes minutes. And then afterwards, we end up with a bunch of different
9:01 campaigns. This is an example of one of them that says for, I don't know, Nick
9:05 just wrote you a blog post. Hey Nick, big fan of your work. Was on one second
9:08 copy site earlier. Thought I'd write you a free 500word sample article with a
9:11 twist. I know giving value up front is how you form connections. So I thought
9:13 I'd start with that. If you have any interest, just reply and I'll send it
9:15 right over. If you want more, just say the word. Got a few other interesting
9:18 pieces of content for you. Uh it also gives you different variants that you
9:21 could use to split test and then also even like a little follow-up. So in that
9:24 way, you know, me just templating out like 80 90% of my work significantly
9:28 streamlines the actual offer building process. And nowadays, because the
9:31 models themselves are getting really good, like I was doing this with Opus
9:35 4.5, now I'm doing it with Opus 4.6. Because the models are getting really
9:38 good, like half the time I don't actually make a change. I just say,
9:41 "Okay, sure. Let's give this a test and we'll see how this performs relative to
9:44 some of the other ones." The last skill I want to show you is a simple website
9:48 builder. Basically, to make a long story short, you know, we do a fair amount of
9:52 outreach. My goal here is I wanted a way to quickly and easily whip up
9:56 highquality, reasonably templated, but also pretty unique websites for clients
10:00 or prospects and then just send them over whether or not I have their
10:03 business. I was thinking about ways to offer more value while reaching out to
10:06 people cold. And um me literally building like endto-end high-quality
10:10 websites and then giving them a link that they could use to access them in
10:13 one shot. It's probably like the biggest value ad that you can get so far. So
10:17 this is like very very lowhanging fruit. And I would also say that there's quite
10:20 uh room right now to do knowledge arbitrage where basically because you
10:24 have access to tools like this and then most prospects don't. You could offer
10:27 either completely free websites or you could offer like a really simple or
10:31 templated website or something like that in order to really blow their socks off.
10:33 When it's done, it actually pushes this to Netlfy, which is a hosting service
10:37 that lets me get a website like this. Then I can actually just click this link
10:40 and then I have the website right over here. And I don't think you guys could
10:44 see it so super clearly, but um it's pretty high quality. you know, it uses
10:48 its own customized prompt in order to build slightly different navigation bars
10:53 and menus and then uh different content types and stuff like that. And then you
10:56 end up with I think what most people consider to be pretty high quality. In
10:58 case you guys are interested, I showcased a really similar builder on my
11:02 Cloud Code course a little while ago and then my Gemini course as well. Um, you
11:05 know, this is pretty high quality, right? I think if an average person were
11:08 to get a website like this for free, they'd probably be like, "No way,
11:11 really. There has to be a catch." But nowadays, you can generate these
11:14 deliverables for basically cents on the dollar and then get like highquality
11:19 animations and stuff like that um with literally you just putting one prompt.
11:22 So obviously I wanted to build a bunch of economically valuable skills and I
11:27 have dozens more probably close to 30 or 40 right now that manage most aspects of
11:31 my business. But um I also wanted to show you guys sorry that's my alarm
11:34 here. I wanted to show you guys how easy it was also just to build skills that
11:38 handle like really nuanced one-off things. And so, um, in addition, I also
11:41 built a couple of cool ones that I want to show you. One called We Work Booking,
11:44 another called Amazon Shopping. And I mean, like, these aren't even really as
11:47 businessy as they are me just like dealing with a problem that I had. So,
11:51 for the Weiwork one, um, We Work is a shared co-working space, which basically
11:55 allows you to book and then show up sometime in the next few hours. Uh, you
11:58 have access to things like kombucha on tap sometimes. Uh, I don't know, free
12:02 coffee. Well, it's not free. I guess you're paying for it. Looks like their
12:05 marketing's worked on me. But to make a long story short, um, in order for me to
12:09 access Weiwork, I need to like book ahead every day. And uh, one day I tried
12:13 booking the same day and then I showed up and my card reader didn't work. And I
12:15 was like, why? And they're like, well, it's usually good to book ahead because
12:18 sometimes the card reader takes times to sync. And I was like, well, this sucks.
12:21 I don't want to freaking do this. And then I realized like, wait a second, I
12:24 have Claude now. So why don't I just write a skill that just automatically
12:27 books me every single day for like the next 30 days? And so that's what I did.
12:31 I basically just took like 30 seconds of my day. Then I said, "Hey, write me a
12:35 skill that books me in for Weiwork." So when I want to access it, I just go,
12:39 "Book me into Weiwork." Then this goes and books me automatically for like the
12:43 next 30 days ahead using some credentials that I've already set up.
12:47 And then it also like actually literally opens up a tab. And so you can see this
12:50 is doing it right here at the one that I currently frequent, um, Steven Avenue
12:54 Place. So I'm just going to say, "Hey, hit me. Give this a book." And then
12:57 it'll go through and actually make like HTTP requests behind the scenes in order
13:01 to push. if necessary, it'll actually also open up Weiiwork. So, you could see
13:05 it actually like do the freaking desk bookings, which is wild. Um, and you
13:09 know, watching this thing go is pretty neat. And then after I built this and
13:12 then, you know, did all of the Weiwork bookings, I also naturally got more
13:15 interested and like, hm, what could I automate with my browser? And I ended up
13:18 creating another skill. So, that was a pretty interesting one-off build. Takes
13:22 just like 30 seconds as you see there. You just have to log into whatever
13:24 service and then say, hey, I want you to automate this with Chrome DevTools MCP.
13:27 I'll show you guys how to do that as well. But I was also kind of interested
13:31 in um what else I could automate that was just kind of like a silly personal
13:34 task that I usually have to do. And one thing that I do nowadays quite a bit is
13:37 like I just buy stuff off Amazon and then I have it shipped directly to my
13:40 door. And I do that because I think, you know, I'm making a fair amount of money
13:42 and I don't really want to drive all the way over to Home Depot or whatever
13:45 because it'll just consume a lot of time and energy and ultimately be
13:48 unnecessary. Why not pay for the premium of having somebody else do it? The only
13:51 issue is, as I'm sure you guys know, anybody here that shops on Amazon,
13:54 there's like a bajillion listings nowadays with like um crappy
14:00 SEO and they try and use every term to rank for everything. So, what I did is I
14:02 just built a really simple Chrome DevTools MCP called Amazon Shopping. And
14:08 what I do is I simply say, "Hey, find me the best floss on Amazon." It then loads
14:14 up the Amazon skill right over here. And then it opens up Amazon, in my case, CA.
14:17 And I'm just going to make that a little bit smaller here so you guys can't see
14:21 my postal code. And then it automates the process of like going through Amazon
14:25 and then looking for whatever the heck I want. So in my case, I'm looking for
14:27 dental floss, right? Well, it's just going to open up one little thing here.
14:30 And then it's going to go through and it's going to compare all of the dental
14:34 floss with each other to find basically the best combination of cost and then
14:38 effectiveness that does whatever I want. And at the end of it here, you can see
14:41 we have all of the products that it's recommending um literally in just like a
14:44 little sheet. So I can give this a quick click. We also have the cost, the
14:49 popularity, the why, buy, and so on and so forth. And, you know, it basically
14:52 uses all these things in order to recommend them. So, I basically built
14:56 like a personal shopper app. Um, I've done the same thing with like grocery
14:58 delivery, which is pretty cool. So, I just one shop my grocery delivery. I uh
15:02 go on, you know, Instacart or another related service, click a few buttons,
15:05 and then boom, you know, I'm basically done for the rest of the year. Uh, and,
15:08 you know, you could build these sorts of one-off systems at this point now with
15:14 Opus 4.6, you know, GPT 5.2, 2 5.3 or Gemini 3.1, whatever model you want to
15:17 use, whether or not you're using Cloud Code or some other platform. Okay, now
15:21 that you guys have seen at least a little bit about what I think like
15:23 actual helpful, useful skills that realistically move the needle, not just
15:27 glorified um you know, personal assistant demos look like, let me show
15:30 you guys how to actually go ahead and then build them. Okay, so first, if it's
15:34 not already abundantly clear, skills are basically the evolution of standard
15:40 operating procedures just for agents. Whereas SOPs and checklists are for
15:46 human beings for the most part, skills are for these new AI agent
15:50 intelligences. And just as such, you wouldn't really just give a checklist over to an agent
15:56 and say do it. You'd have to translate its language just a little bit into sort
16:00 of like the native format that the agent understands. So, a brief example here.
16:05 If I have maybe a checklist and it's make PB and J sandwich and you know I work at a
16:13 company where all we do all day is make PB&J sandwiches. Boy, would I love to
16:17 live there. You can imagine how new hires go their very first day and then
16:22 they're given a checklist of tasks. And it's basically take bread
16:32 out of basket, put jam on one side, put peanut butter on the other side, and
16:37 then maybe one final one that just says stick them together. So this might be
16:41 for a human. And essentially, as a human goes through it, what do we do? We
16:45 obviously check all of these tasks off. That's just how standard operating
16:48 procedures are in business. They just for the most part standardize things,
16:50 hence the name. Because this is organized a little bit differently for
16:54 agents than, you know, for human beings, we organize them into what are called
16:59 markdown or MD files. And these markdown files are basically just text files with
17:03 a little bit of a twist. For instance, I have a markdown or MD file open over
17:06 here. If I try and create a new file here, and then I call it skillspec.txt,
17:11 and then I paste everything in, you notice how everything is the same color.
17:16 Whereas, if I go back to that markdown variant, we all of a sudden have blue,
17:20 we have orange, we even have some green. Well, the way that markdown files work
17:24 is they take base data and then they also take a little bit of formatting and
17:28 then they apply the formatting to the base data just so that it's a little bit
17:31 easier to see, a little bit more interpretable. And so if you guys could
17:36 tell, the data that we're marking up here, okay, funnily because it's called
17:39 markdown, but the data that we're changing the the color of and stuff like
17:43 that is usually prepended with a couple of symbols. The first symbol here is
17:47 this little hashtag. This always refers to like a header. And
17:51 so you can kind of think of it like a chapter heading in a book. Another one
17:55 are these sort of back ticks here, which typically refer to files. There are a
17:59 couple of other little tiny formatting things here like links and so on and so
18:03 forth that you don't really get with txt files, but he'll hopefully you guys
18:07 understand that it's not really that big of a deal to memorize everything in a
18:11 markdown file. You just have to know that they're slightly different
18:13 formatting compared to text files. The reason why is because you're never
18:16 really going to have to make them yourself. All you're going to have to do
18:19 is just give your agent some bullet points, usually a brief description,
18:22 which I'll show you how to do, and then it'll combine all of that into the
18:27 optimal markdown skill format for you. Now, unlike just a few months ago, today
18:33 every major provider on Earth basically, so our open AIs, our Googles, our I
18:37 wrote Claude here, but I really meant anthropic. They're all moving towards
18:42 harnessing and then making skills generally available just as part of like
18:47 their their their LLMs. And so here you could see Anthropic has a page on their
18:52 claude code docs that say extend claude with skills. These are the guys that
18:55 sort of made skills up first. But because of popularity, a bunch of people
19:00 um that were openi developers and stuff like that started talking about it and
19:02 asking for it. And so they've created their own skills format. It's very very
19:08 close to the cloud code skills format. And then also over here the Gemini CLI,
19:13 you can see you can add agent skills. Uh basically no matter what framework or
19:18 LLM or platform you're using, you can do the exact same thing. So my rule of
19:22 thumb is if you guys have a standard operating procedure, if you have an SOP,
19:27 you have a skill. Okay? All you really have to do is just copy that SOP from
19:32 your business and feed it into Claude or Gemini or Codeex and ask it to make a
19:37 skill. As of the time of this recording, the skills spec or definition is uh it's
19:44 not like baked directly into the models. So the one thing that you have to do in
19:47 addition to that is you basically have to give it the skill spec and then ask
19:52 it to turn your task list into a skill. But assuming you can, you know, drag a
19:56 file into your freaking uh working directory and assuming you have any SOPs
20:00 at all, even poorly written ones, you now basically have an agent capable of
20:05 doing a big chunk of your knowledge work. Once you're done with this, you
20:08 just ask the agent to run it at least once cuz you have to test it end to end.
20:12 On the first run, if the agent needs something else, it'll ask you for it.
20:15 Otherwise, it'll build the asset yourself itself. So maybe like an Excel
20:19 file or something like that, it'll actually put that together for you.
20:22 Although obviously if you have assets like some context or PDFs or whatever,
20:26 you can provide them no problem. And then assuming that you've prompted it
20:30 right with the spec that I'm going to provide you, every time the skill runs
20:34 after that, if the agent finds a mistake while doing it or maybe a service is out
20:37 or it doesn't have the knowledge it needs, it'll automatically figure out
20:41 how to solve it and then it'll patch the skill for you. And so in that way,
20:44 here's the real power of skills. They're self- annealing over time. They heal
20:48 themselves. They get better and they improve constantly. Much like an
20:52 ambitious, intelligent staff member who sees a checklist, notices that there's a
20:56 gap, and then chooses to fill it, skills are the same thing, just with agents,
20:59 which is what makes them so incredible. So, why don't we get started and produce
21:04 ourselves some skills and then as we produce, and then shortly after, I'll
21:08 also run through the special directory of Claude/skills, which uh with Claude is how you organize
21:15 these things. I'll also touch on just a couple of other ones in case you're
21:18 using a different LLM, but for the most part, they're going to be exactly the
21:21 same. So, the very first thing you need to do is you need to grab the
21:24 skillspec.md. And you can grab that down below from the Google Drive link. Um, it's free as
21:29 mentioned. You don't need to sign up or give me your email. I don't need it. I
21:33 have enough of those. And what you're going to see in here is basically the
21:38 condensed spec or definition of skills as well as the file structure, all the
21:43 locations, all the different formatting options and stuff like that. Um, and
21:47 what we're going to do is we're going to give this to our agent alongside our
21:50 request so that it can just figure out how to put together the highest quality
21:54 skills for us. What I did in order to put this together, by the way, is I just
21:59 went over to Enthropics Cloud Code docs with skills page. I copied the entire
22:03 thing, fed it back into the LLM, and then just said, "Hey, I want you to
22:06 really, really shorten this, compress this, significantly increase the
22:10 information density because I don't want this to be 500 lines. I want this to be,
22:14 you know, 150 to 200." Okay. So, once we have this, basically what we do is we
22:18 just make this our claw.md, you know, just so we're on the same page
22:21 here. Cloud.md is just the prompt that's injected at the top of every new
22:25 conversation. And so, what I'm going to do is I'm just going to rename this and
22:29 just call this oldcloud.md. And this one here is going to be the new claw.md or
22:34 just claw.md. Any file in your root directory that contains the capital c a
22:40 c l a u d e.lorcase md will automatically be prepended and added to
22:44 the prompt. And so what we do is basically by by doing this every
22:48 workspace we're in is just automatically going to know what skills are. If you
22:51 have other stuff in your cloudmd you can also just copy this and then paste this
22:55 um as well. Although, you know, at least for the purposes of creating skills, I
22:58 recommend just being as short and and straightforward and basically having
23:01 this in as few lines as humanly possible. Next up, to build your skill,
23:05 just open up a new agent window. So, I've just done that here with Claude
23:08 Code. And then just tell it what you want it to do. I'm going to use a voice
23:11 transcription tool, Whisper Flow, just to put in my own SOP. What I want to do,
23:16 at least for this demo, is just build a simple inbox cleanup tool that just goes
23:21 through my inbox. It then opens up all of the um unread messages and then just
23:26 uh immediately reads all the ones that are pointless and sort of I don't know
23:30 invoice reminders or follow-ups about whatever or simple templates and then I
23:34 want it to highlight um like realistically the high priority emails
23:38 in my inbox. So I just have a short list of everything. So that's pretty simple
23:40 and straightforward. I'm just going to hold this little key down and then say,
23:46 I'd like to build a Claude code skill that goes through my inbox, identifies
23:53 all unread messages, reads all of the unread messages, and then marks as read
23:59 all of the ones that aren't uh important to me or high priority. My definition
24:04 for important is basically anything that isn't automatically generated. If the
24:10 email is automatically generated by like a service or it's a simple notification
24:14 email or whatever, it's not important. If the email is written by like some
24:18 cold email copywriting fellow who's trying to reach out and get my business,
24:22 it's not important. The only stuff that is important are emails that are
24:27 personalized, customized, and that give off the vibe that they were written
24:31 basically just for me. Once we're done, we're going to feed that right in. Then
24:34 we're going to go ahead and build it. The very first thing it's going to do is
24:38 look at all of the existing skills and any Gmail code inside of my workspace to
24:42 see if there's anything that it could use. And it's doing this because, you
24:46 know, as human beings go, we don't really want to rebuild the wheel if we
24:50 don't have to, right? So, this is it just reading everything right now. You
24:54 can see that it's also changed to plan mode. That's why that little thing in
24:57 the bottom lefthand corner is changed alongside that red bypass permissions
25:02 color to blue. This happens more or less anytime you try and build something that
25:05 it would consider to be even slightly complicated. It wants to go into plan
25:09 mode so that it can plan out the build of the skill. Now, it correctly went
25:12 through and found that I had a couple of configuration files that allowed it to
25:16 use different emails here. And for the purposes of this demo, I'm just going to
25:20 use this one. Obviously, if you didn't have this, it would just ask you what
25:23 email do you want to set up. Now, what it's going to do is actually create the
25:27 skill.mmd. It's also going to create a script for me which uh looks like it's
25:32 going to call inbox_cleaner. Then afterwards it's going to test it
25:35 end to end. Why? Because that's just what's in our spec sheet. We basically
25:38 tell it how to do all of this stuff natively. Now what it's going to do is
25:43 create the script. You can see that because I have a pre-existing skill,
25:46 that follow-up one that I was showing you guys, it gets to borrow based off of
25:50 that. And you can see it's actually gone ahead and wrote the script. If I go to
25:54 inbox_cleaner here and actually look at the skill.md, you can see it right over here. If we
26:01 just make this a little bit bigger, let's move this here so we could focus
26:06 on this. Basically, what we have is a stepbystep guide for a future agent on
26:11 how to do the task that I just asked it to. And you'll see that it's now been
26:15 organized in a very particular way. For instance, there are these three dashes
26:19 up at the top, three dashes underneath the section. There's a name key that
26:24 says inbox cleaner, a description key that says clean up Gmail inbox by
26:27 reading all unread emails using AI to identify which ones are generally
26:30 important and marking the rest as red. Using cleaning inbox, triaging email or
26:34 clearing unread notifications. Then it also allows it to use some tools. Now,
26:37 you may be wondering, you know, I think these two are pretty self-explanatory,
26:41 but why is it allowing certain tools? Well, Claude and all other LLMs
26:45 basically have access to some little pieces of software under the hood. Like
26:50 they can read files, they can GP, glob or bash. And these are just various like
26:54 command line utilities that allow it to look through files, look for specific
26:58 sections and so on and so forth. And so we've done with this skill, okay, it's
27:01 almost like a sub agent. We've basically given it the ability to do these sorts
27:05 of things when we invoke or instantiate this. Aside from that, we have the
27:10 title, we have the goal, what counts as important, what gets marked as red.
27:13 Notice how it basically took everything that I have provided it in just a few
27:18 lines of voice transcription and then turn this into a pretty coherent and
27:22 then consistent SOP. Now, going back to the actual building thread here, it's
27:26 actually gone ahead, wrote the script, and now it's doing the classification.
27:29 If we give this button a quick little click, you could see that it's now doing
27:33 it in batches of 10, which I didn't even ask it to. Okay, it's identified which
27:37 ones it considers as important, which ones are not important, and now it's
27:40 saying, "Hey, do you want me to mark these 97 emails as red?" Now, I don't
27:45 actually want it to mark the 97 as red. What I wanted to do is I wanted to show
27:48 me what they are so I could tell it which ones are right. So, I'll say, "No,
27:52 I'd actually like you to show me all of the 97 that you marked as red so that I
27:57 could take a look and tell you if there are any exceptions." Now, what it's
28:00 going to do is give me a big list of all of the emails. And as you can see here,
28:06 it has correctly identified that 79 of the emails I got were from a broken make
28:10 scenario. So that's pretty interesting. Um, obviously these are more or less
28:13 exactly the sorts of emails I don't really want to. One is some normal
28:18 commission notification from referral stack. Another is a subscription renewal
28:22 for that voice transcription flow. Another's from Bright Data. Another's
28:25 from whatever company. I have a bunch of cold outreach over here. Okay. I have
28:31 sponsorship pitches. oneword yes reply to lead magnet. I guess what I'm trying
28:35 to say is this did a really good job. There isn't a single email here that I
28:39 actually really care about. So I'm just going to say nope mark all as red.
28:44 That's pretty sweet. Right now at any point in time I could open up a new claw
28:48 instance. And we have to go new just because of how skills work. Okay. Then
28:53 we can go /inbox cleaner and then we can run it. And because we are running based
28:58 off of this SOP, it's always going to run in the exact same way. It's going to
29:01 be repeatable, consistent, and dependable. Also, if it runs into any
29:05 issue, let's say the script doesn't work, let's say there's an API rate
29:09 limit or something like that. It will identify the problem and then
29:12 automatically go through, fix it, and then rewrite its own skill. In my case,
29:16 I've got a lot of unread emails. So, I think what I'll do is I'll say, "Hey,
29:20 I've got a ton of unreads. Could you actually go through and fetch a thousand
29:24 instead of 100?" And now I have the ability also to modify the way that this
29:28 skill works just a little bit so that instead of the base 100, maybe I can do
29:32 a thousand. Maybe I could pass in a flag like dash a and that stands for amount
29:36 and then I can do you can see that's actually what's going on under the hood.
29:40 It's passing the the a number 1,000 to said script. Now that you guys see how
29:44 these skills actually work in practice and how to easy it is to build one as
29:48 long as you have the right spec sheet. I just want to really quickly deconstruct
29:52 the various parts of this file to make sure that you guys understand how all of
29:56 this stuff works and what each of these parts are for. Okay, so you guys have
30:00 probably already noticed these three little dots here are really different
30:04 from all the formatting elsewhere. What this basically is in at least markdown
30:11 speak is something called front matter. Front matter is a highly optimized way
30:16 of basically putting a tiny summary at the beginning of any markdown file.
30:20 Remember how there's different fields and colors like keys, headings, there's sort of more headings.
30:27 I mean, there's a lot of different things here. Well, you can just think of
30:30 front matter as being the same idea. It's just a different like variable or
30:34 type. And the way that you start it is with these three dashes and then you end
30:39 it with these three dashes. And so this little summary is basically doing you a
30:44 massive service by significantly reducing the number of tokens that you
30:48 need to load into your context window at any point in time. If you think about
30:51 it, what we could be doing every time we load this skill in, if I just copy this
30:55 whole thing and go to a free word counting service, is we could be adding
31:00 373 words, which is approximately 500 tokens into context. Okay. But with
31:04 front matter what we do is instead of storing the entire script into context
31:08 and making it accessible to the model, what we do is we store a much smaller
31:12 summary which is closer to like 60 or 70 tokens and then we say hey if the user
31:17 calls inbox cleaner specifically then and only then do I want you to load the
31:23 other uh you know 350 words or so. Okay? And so what this means is in practice,
31:27 if I were to show you guys all of the different skills that are currently
31:31 loaded into context, they basically all look like this. And so my prompt at the
31:35 very beginning, just assuming I have these three skills, is always basically
31:42 going to be my claude.md up here. Underneath we're going to have our
31:46 skills. So there's going to be one, two, three. And then underneath it'll be your
31:50 prompt, which is basically whatever the heck you want to say. Hey, I want to do
31:55 XYZ. And so this is more or less the structure that every conversation with
32:00 Claude in my case, but obviously any LLM in yours, whether it's Gemini.md,
32:04 agents.mmd, so on and so forth, is going to realistically look like. You'll have
32:08 your prompt down here, but before then, you'll have a bunch of hidden stuff. And
32:11 then what's cool is, you know, if you say, "Hey, I want to do um follow-ups."
32:15 What it's going to do is it's going to say follow-ups. Wait a second. I know
32:19 that we have some follow-ups mentioned earlier on under the skills section.
32:23 Okay, great. That means this must be a standardized skill. And then it'll go
32:26 and actually load the whole follow-up nurture directly into its context, which
32:31 will be the rest of this. Then it'll basically stick the rest right here and
32:34 then resubmit the prompt. And now it knows everything that it needs to do in
32:37 order to obviously get that right. So the agent would go, "Okay, great. I have
32:43 the skill spec right here. Let me follow each step." And then it'll go and you
32:47 know start with number one which is load pending leads number two for each
32:51 pending lead and so on and so on and so forth. Now this actually has a term this
32:56 is called skill matching. Anytime a user types a request like hey get me 50
33:01 dental leads. It'll automatically scan those descriptions or that front matter
33:05 for whatever the most relevant description is. It says find business
33:10 leads here. It says find business leads here. It'll say oh this must be the one
33:13 that they want. Then it'll load the full skill.mmd into context before finally
33:18 calling some scripts. Okay. And then giving you guys an output. And where you
33:22 put those scripts don't really matter. The basic skill definition, I think,
33:25 just bundles it all into one folder. You could put the scripts somewhere else if
33:28 you want to keep them organized. And if you have other assets like PDFs or
33:31 whatever, you can build whatever scheme you want to keep those files handy. The
33:34 reason we do this, of course, is both for efficiency because the less context
33:39 in a model's context window at any mo point in time, the higher the quality of
33:43 the output, but also for costs. Providers like Enthropic and so on and
33:46 so forth actually don't want you to spend an arm and a leg, believe it or
33:50 not. Um, they want you to continue using their software, and in order to do so,
33:53 they need to make it like reasonably costefficient for you and deliver a high
33:58 ROI to you um at the same time. And so this concept of progressive disclosure
34:02 is now not only really being applied to skills but a lot of other things as
34:05 well. And all this is again in the pursuit of reduced cost, reduced
34:09 overhead. Okay, let's build three more skills with what we know now. And just
34:12 for simplicity and to make this as fast as humanly possible, both for myself and
34:16 for you guys, what I'm going to do is I'm going to open up three separate
34:21 Claude Code little plug-in widgets here. And in the first I'm going to switch all
34:26 these to bypass permissions and make this a little bit smaller so
34:30 you guys could see. And then in the first I'm going to paste in skill number
34:34 one. In the second I'll paste in this skill number two. And the third I'll
34:38 paste in the skill number three. What I'm going to say is create skill one,
34:43 create skill two, and create skill three. And we're just going to trigger
34:45 all three of these and see how they operate. So, what skill one is is it's a
34:50 simple way to turn meeting notes to action items. We basically just paste in
34:54 a meeting transcript and get a structured action item with owners and
34:58 deadlines and so on and so forth. The idea here is to show you guys that a
35:01 skill can be really nothing more than a well-written SOP, which doesn't even
35:05 include Python scripts or or anything like that. The second one is an invoice
35:09 data extractor, which is basically going to be skill.mmd plus some Python script.
35:13 So, we're going to take a PDF invoice, then return structured JSON in the form
35:17 of a vendor, amount, date, line items, and tax. And this is to help you guys
35:21 see how skills can delegate deterministic work to code. Aka, we
35:24 could use Python scripts and whatnot, like you guys have already seen. And
35:27 then the third skill is going to be a content repurposer, which is going to
35:30 let us take a transcript and then output a tweet thread, a LinkedIn post, and
35:34 then a newsletter draft, all in parallel. And the idea here is we're
35:37 going to have templates. I'll show you guys some brief examples and I'll also
35:41 show you guys how this works with some instructions. And so before I was even
35:44 finished reading this freaking thing, we're almost already done two out of the
35:48 three. Isn't that wild? Anyway, the first skill, pretty straightforward,
35:51 just called meeting notes. It just processes meeting notes. So, what we can
35:55 do here if we want is we could say, "Hey, I want you to output specific
35:59 elements of the meeting." In this case, it has some decisions, open questions,
36:03 and so on and so forth. But I also want you to find and identify any contact
36:08 information. So I'll say, "Hey, this is great. I want you to update the skill
36:12 also to extract any contact information that's mentioned. I'm looking for things
36:16 like email addresses, websites. I'm looking for phone numbers, and so on and
36:20 so forth. Anything that somebody offhand mentions during the call. Sometimes we
36:24 also specifically ask for this information for follow-up reasons. So it
36:27 would be nice to have it in a structured sort of way. Over here, it's now created
36:31 both files of the invoice data extractor. So we can see there's a
36:34 skill.m MD here which basically takes the PDF invoice return structure JSON
36:38 with vendor. So in order for me to really appreciate how this works, I'm
36:41 actually going to go and grab myself a brief little invoice. Uh just feed it in
36:45 an example and then we'll see how it works. I see a couple of services here
36:48 that have sample invoices. This one looks pretty good. Why don't I download
36:53 this? I'll say example_invoice. What I want to do is I just want to feed
36:56 this in right over here. And I'll say great, use this on example_invoice.pdf.
37:01 PDF. Then over here, it looks like this is now done. And what I'm going to do is
37:05 I'm going to go find some of my own content, just feed that in just as an
37:08 example. So, we'll go on my YouTube channel and I'll find something that is
37:12 probably not 4 hours because that would take a lot of tokens for no purpose.
37:16 This 10-minute one looks pretty good. >> Hey, sir. >> Let's get my fat ugly head to shut the
37:21 heck up. I'll scroll down here to where it says show transcript. And then right
37:25 now, just cuz it's an example, I'm just going to copy all the stuff in manually.
37:27 But obviously, you guys could do whatever you want with this. Going back
37:31 here, I'll then paste this in. And what I want to do is I basically just want to
37:35 see whether or not I can do all of those three things. I'm also going to take a
37:39 quick peek here at the script. And um this is what I wanted to happen
37:42 actually, which is nice. It ran into an issue with one of the summary rows and
37:46 the subtotal. So what it's doing now is it's actually going to fix the script's
37:50 filtering logic so that it doesn't catch the subtotal or the credit terms or
37:54 something like that. Okay, cool. On the right hand side, it's extracted this
37:57 transcript. Looks pretty good. And now what it's going to do is spawn all three
38:01 formats in parallel, which is its way for saying it's going to rewrite my uh
38:04 main transcript and turn this into like blog posts and Twitter threads, whatever
38:07 the heck it's going to do. While all of this is happening, this is me pictured.
38:12 I'm just manifesting all of the skills. I don't actually have to do too much. I
38:16 just kind of close my eyes and wait and then they appear. Taking a look at this
38:19 tweet thread over here. Let's just make this look a little bit prettier. We now
38:24 have a bunch of tweets that are divided. This is automatic, which is pretty cool.
38:27 A 16-year-old posted a prompt on Twitter. Former professional web
38:30 designer copy pasted into Gemini. The A produced a better site than he could in
38:34 5 hours. In minutes, design is commoditized. Sure, you guys have seen
38:37 these sorts of threads all over the internet. Let me tell you why. It's
38:41 because people have built very simple repurposing pipelines like this. And
38:44 although that one just took me 5 seconds, the content's reasonable. Not
38:47 going to say it's going to win any awards. It's pretty LLM. But hopefully
38:52 you guys could see with uh at least a few neurons between your two ears, you
38:55 guys could repurpose, change the tone of voice and so on and so forth and make it
38:58 something actually pretty cool. Likewise, we also have LinkedIn posts
39:02 and then we even have a newsletter draft for me. So that's pretty sweet. You can
39:06 see that its design decision was to do this via sonnet sub agents. You guys can
39:10 modify this SOP at any point with literally just a line. Finally, in the
39:14 middle here, skill 2 is all good to go. Great job. It's gone through and it's
39:17 actually updated the main skill as well, which is quite neat. If you guys were
39:20 paying attention, you would have noted that every time that we add a new skill
39:25 or change skills or whatever, um there's a top left folder over here that's
39:29 claude/skills that changes. And this is essentially the format and file directory structure
39:36 of a skill. So, it's worth us pointing it out. Now, in case you guys have never
39:39 seen this convention, in the top leftmost corner, you see how there's a
39:43 little dot in front of the claw folder? Well, that's basically just the kind of
39:47 universal file explorer convention for hidden. Now, if I open up this exact
39:52 folder, the skills example folder in my Mac OS's native file explorer, you
39:57 notice that it looks pretty different from what we're seeing over here. Notice
40:00 how on the left hand side we haveclaw.tmp, then we have 1 second copy config data.
40:05 And notice over here that the list actually starts at 1 second copy, then
40:09 goes claw.md, then it goes config, and then it goes data. So, I mean, aside
40:12 from the fact that this is organized a little bit differently. Generally
40:14 speaking, we always have folders organized first in the file explorer
40:17 here. So, if we kind of match that up one for one, you'll notice that the only
40:20 things that are missing are basically the files and the folders with little
40:24 periods in front of them. So, the first thing to note is like most people just
40:27 have no idea where this thing is because when they go to the file explorer, like
40:30 it's always hidden. Well, in Mac OS, it's pretty easy to reveal. You just go
40:35 shift command uh dot period basically and then you can show all hidden
40:38 folders. They'll be present. It's just they'll be a little bit um more
40:41 translucent and then we can actually open it up and then go through all the
40:45 skills. Okay, so that's the very first thing to know. On Windows, which you
40:48 guys may be using, it'll be a different hotkey. Just Google it. There's uh
40:50 plenty of information on there about hidden files. I think you can also just
40:53 rightclick and go like show hidden files or something cuz I used to have a PC and
40:56 that was pretty straightforward. But uh yeah, essentially inside of every main
41:01 work space in any sort of IDE, if you do have skills, they're probably going to
41:03 be hidden from you, which is kind of difficult and intimidating to to kind of
41:08 understand. Okay. So, and then after that, the way that it's organized, which
41:12 is not super plainly clear here, so to make it a little bit clear for you guys
41:15 is you have the claude folder, right? And then underneath the dotcloud folder,
41:20 you have the skills folder. Then underneath the skills folder, you
41:23 actually have the specific skill folder that you want. So, let's just go, you
41:28 know, inbox cleaner. That's another folder. And then inside of that folder, you actually have the
41:34 skill file itself. So, then you have the skill.md. So notice how nested this is.
41:38 I mean it's a massive pain in the butt. I don't know why they do it like this,
41:41 but um they make it pretty difficult for most beginners to understand. And they
41:44 do so in this sort of like hierarchy that begins again with claude, then go
41:47 skills, then goes in, you know, the specific name of the skill and then the
41:51 skill MD itself. And then over here, okay, and this is sort of optional, but
41:54 this is how anthropic typically does it. Here you can put your other stuff.
41:59 What is your other stuff? Well, your other stuff is going to be things like
42:02 your scripts. So, hypothetically, if you had like scripts that helped you do the
42:06 invoice cleaner, inbox cleaner or whatever, maybe an inbox uh- cleaner.py,
42:10 you know, you'd probably put it somewhere in here. I personally like
42:13 doing it in a different folder called execution sumi. I just think it makes
42:17 more sense, a little bit cleaner. Uh likewise, if you have, I don't know,
42:21 like your Gmail off tokens or something, you know, you might include them over
42:24 here. If you have, I don't know, some other file like example.txt,
42:29 um you'd include it over here. And so this is just like the base way that all
42:32 skills are organized. And yeah, I think it's just unfortunate that it's usually
42:34 in a hidden folder, which is kind of intimidating. I should note that the
42:37 skills live right alongside a bunch of other hidden folders like agents and
42:40 stuff like that. I'm not going to talk about those now, but if you want more
42:42 context on that, just check out my full length Cloud Code course. I got a ton of
42:45 juice for you in there. Okay, last thing I really want to do is just talk a
42:49 little bit about what skills are actually worth making. You know, I do a
42:54 lot of business consulting now. uh somebody that started in door-to-d
42:58 dooror sales, I saw the importance of focusing on the levers that actually
43:02 increased revenue first and foremost above all else. And so for me, it was
43:05 always doortodoor. But then transitioning into marketing and then
43:10 broader through things like uh content creation, articles, and now videos, I
43:14 think a lot of people are very misaligned on what like a high revenue
43:19 SOP even looks like to begin with. And the reality is with agents nowadays, you
43:24 can do anything, but that doesn't mean that you should do everything. These
43:29 things are so damn good, they can of course design whatever crazy back-end
43:34 fulfillment pipeline you want. But I find most of the demos on the internet
43:37 that talk about stuff like that really miss the mark because we've been able to
43:43 design these pipelines forever and most people that do these sorts of things
43:46 still don't make any money. the difference between people that use new
43:51 upcoming and trendbreaking technology to make money and actually have big, you
43:55 know, success outcomes versus people that play shiny object syndrome all day
44:00 and just collect a bunch of Pokémon cards. Our the former group focuses
44:04 primarily on the front end of their business. So, they focus on the
44:07 acquisition of opportunities. They focus on the generation of leads. They focus
44:12 on mechanisms that improve the quality of the sales experience for their leads.
44:18 They focus on selling closing mechanisms and so on and so forth. And then the
44:22 latter people just overcomplexify the hell out of everything and then make
44:25 every excuse under the sun not to focus on sales and marketing. So this is all
44:29 just like a pretty meandering and long way of saying if you're going to design
44:32 skills, I highly recommend you focus the design of those skills primarily on
44:36 front-end tasks. Like most of the people that are going to be watching this
44:39 video, you know, they're probably going to have some sort of side hustle going
44:42 on or they're going to be looking to get into that realm. The simplest and
44:46 easiest way that you can move the needle in your own business is if you're going
44:49 to design skills and stuff like that, spend your time designing them on things
44:53 that help get you sales, on things that help uh, you know, you do your marketing
44:57 faster and better. And then don't just have them sit around in your folder.
45:00 Actually use them. You know, when I was growing my business, I was doing
45:04 somewhere between 50 to 100 cold calls every single day. When I was going door
45:08 to door, it was something like 80 knocks on physical doors per day. I wish I had
45:12 a hundth of the technology we have today. and I'm not even that old uh to
45:16 go back and then automate that process for me. I don't. But it blows my mind
45:21 how many people do and then they choose not to use it on things that actually
45:24 make money like those laborious doortodoor sessions or or cold calls or
45:29 whatever. Okay, so ultimately things that I have seen work really really well
45:33 content repurposing pipelines. These genuinely can significantly improve your
45:37 reach assuming you're growing a brand. That's for inbound lead generation
45:41 systems. Things that do stuff like scrape leads for you. Those are really
45:45 big skills that use Chrome DevTools MCP to post on forums automatically for you.
45:50 Uh to, you know, do product research quickly and then tell you which threads
45:53 are saying bad things about your product. These things are valuable.
45:57 Skills that, you know, quickly go through a bunch of books and identify
46:00 which things are business expenses and can be written off versus things that
46:04 aren't. These sorts of skills are valuable. The skills that are not super
46:08 valuable are like skills that help you build other skills or skills that uh
46:13 invent some new design framework that assists you in the building of uh AGI.
46:18 Right? These things are not ultimately the stuff that I want you to take away
46:22 from this video. I want you to focus on real practical use cases that actually
46:25 move the needle for your own business or the businesses that you're working with.
46:28 Because at the end of the day, that's the stuff that's going to make the
46:30 economy more efficient. That's the stuff that's going to make you more money as
46:34 uh you know, AI swallows up more and more knowledge work. All right, so
46:37 hopefully I didn't bore anybody to death. Um guys, I'd like to make a big
46:41 ask. If you guys like my content, please subscribe. I realized the other day that
46:46 69 something% of all of the people that regularly watch my content are not
46:49 subscribed. That means about two out of the three people that I'm talking to
46:53 right now, for whatever reason, haven't. If you like the sort of content that I
46:57 make, um, you know, the algorithm really does like it if you subscribe and then
47:00 it pushes my YouTube channel to a bunch of new audiences, but if I don't get
47:03 enough subscribers, then none of that stuff happens and then most of my videos
47:08 bomb. So, I really appreciate anybody that has taken the time and energy to
47:11 subscribe, and I'm not going to do a big cry session here. Although, as a newbie
47:14 YouTuber, I always wondered why everybody was pitching me on subscribe
47:17 this, subscribe that, and now I think I get it. Most people just never do. Um,
47:21 but yeah, I would really appreciate it. Aside from that, um, if you guys like
47:23 this sort of thing, just leave a comment down below with any sort of, uh,
47:27 question or recommendation as to future content, and I will happily take that
47:31 into consideration. I have skills that now scan my YouTube comments and then
47:34 use that to help me ideate new videos to make, and so on and so forth. So, yeah,
47:38 a lot of cool stuff in the works for you guys, but I'd definitely love to know
47:41 what direction to take it. Thank you very much for watching. Have a lovely
$

How to Build Claude Skills that Generate Revenue (Full Course)

@nicksaraev 47:44 11 chapters
[AI agents and automation][revenue model and pricing strategy][content creation and YouTube][developer tools and coding][productivity and workflows]
// chapters
// description

🔥 Join Maker School & get customer #1 guaranteed: https://skool.com/makerschool/about 📚 Watch my NEW 2026 Claude Code course: https://www.youtube.com/watch?v=QoQBzR1NIqI 💎 Free Skills: https://drive.google.com/drive/folders/1YWXOAvhDFn1ELgYywYaZpOE_N65THmzT 📚 Free multi-hour courses → Claude Code (4hr full course): https://www.youtube.com/watch?v=QoQBzR1NIqI → Vibe Coding w/ Antigravity (6hr full course): https://www.youtube.com/watch?v=gcuR_-rzlDw → Agentic Workflows (6hr full course): https:/

now: 0:00
// tags
[AI agents and automation][revenue model and pricing strategy][content creation and YouTube][developer tools and coding][productivity and workflows]