# Batch 4: page 9, the book's purpose, and the team's pace > Batch 4: page 9, the book's purpose, and the team's pace: one of the editor's briefs of 2026-10-31__rc1-paper-review, published in full as given. *Source: · site v0.1.207 · generated from the same content as the page, so the two cannot drift.* --- The editor, in the conductor's session, 31 October 2026. Kept as he gave it. > Okay, I'm on page 9 now. This is a good example of an image that has a wrong size. And actually, my question is, have any agent picked it up, right? Because it's quite a big bug. So if we haven't, you really need to figure out what's missing on the workflow, because that's a perfect example of an image that is massive, right, compared to the rest of the page. And also, this is a good example of, you know, that 10%, 10 agents, is probably not relevant, right? Like, you know, we need to understand what we're trying to communicate. Because I think with every page, we should also be asking what we're trying to say here, what we're trying to communicate, what's the objective, and logistical things. And like, what we want is communicate ideas. You know, we want to get the user to do a paradigm shift. I think, in fact, even, you know, one of the things that I feel we got right with the founders book, and by the way, we can grab some content from it, right? There's nothing wrong with bringing some content from it here, but one of the things I think we got right in the end was this idea that that was a book where the user was going to learn on the way. So it's almost like we have a workflow here, and our job is to get the user to replicate this workflow. So it's almost like every single chapter, every single thing should have prompts, examples, case studies, you know, like do this now, do that now, you know, like create a Claude account, create a Teams account, create a Workspace account, connect that to your thing, give it this A-B policy, you know, give it this prompt, have this interview, you know, try this out, create this. So it's almost like as we introduce the concepts of what we use and how it works, we should be helping the user, right? We should be guiding the user to replicate. So it's almost like the user by the end of the book should be able to replicate everything we created. And I think this is kind of the point of the book, right? Is to teach. Because if you think about it, I want this book that I can give to the people I work with and say, right, now replicate this, right? So now I'm replicating this workflow left, right, and center because I know how to do it. It's easy to onboard. It's easy to point. So we should be telling people to follow it up, even if in the beginning is to say, which I, sometimes I say these days, I say, look, point your Claude to this repo and ask it these questions. And that's probably a good way to do it, right? You know, is to say, hey, here's a repo that has information, point Claude to it, or here's a section that has information, or here's a QR code, give this to your Claude, right? And I think, you know, this idea of making the book super practical for the users, where at every section we have these examples and these infographics and these mini animations or in the book format, and we show Claude small and we show like how it works, and we show the arrow, you are here, you're going to put the data here, you're going to get this, you're going to get an agent to do this, and you're going to set up a schedule, and this is how the schedule is going to work, and then you're going to have the conductor doing this, and then so it's kind of like it, and actually probably it's a good idea, right? So probably like the first agent that the they should do is the conductor, right? And maybe we guide them there. Okay, so what you want to do is you want to have a dedicated Claude account for this. Create a team account, it costs you X, and now you have a clean stuff. So then we start to say, okay, this is using Claude Code, this is using Claude Thing. Now you create a GitHub account. So it's almost like these prerequisites that we can do. And this is where the augmented sort of page is quite interesting, because what we can do is we can have super detailed pages in our website for this kind of stuff. But on the book, we can say, okay, you need to create a GitHub account, right? And maybe put some screenshots, collage of the in guidance of the workflow, and then says, go to this QR code, click here, right? And actually, QR codes is better than links, right? Because QR codes are actually actionable links on the page, right? So, versus a link, which is in the book, doesn't mean anything, right? So you go to QR code, open that up, and that basically loads up the page with more instructions. And then you can have more or less instructions of how to do something. So I think this is kind of important, this idea that we are helping the user to be at the same point I have. In fact, maybe that's the objective here. The objective is we want the reader by the end of the book to be where I am at the start of the book, right? And I think that's important. And you can personalize this. You can say, look, that's my objective in the book. My objective here is to get you because you are either somebody I work with or somebody that I try to help or somebody that would like to replicate the stuff that I do, right? So that's, and in this case, this is what I achieved, right? So I created this. So this is why, for example, an area like create your own presentation UI, that's a whole section, because if you think about it, that's the idea. The idea here is that I'm going, so in a way, if you think about it, I'm showing how I did it. I'm showing the UI that I created for me and the features that it has and why it's so powerful and why I added those features and from a design point of view, why it works and why it doesn't work. But that is about them creating their own version. Because that's what we want to show. We want to show that, in a way, the power that I have is I now don't separate form from function. I don't separate the tool from the creation. I build the tool as I create, right? And then, of course, I build the tools in a scalable way so that I get to reuse them. So then next time, I don't have to build as much. So I was only able to do, again, this presentation in this short time because I already had figured out 20, 30, 40 different things. And that's what we want the reader to do. We want we can't sometimes give all it to the reader. We need to give it bit by bit, right? And then everything in the end comes together. So I think we can show the end result. We can show why it works. We can show the complexity. We can show that, look, I was able to get 30 agents co-collaborating, but you don't start there. You start with one. And again, if you seek in my history, I've been doing baby steps along the way until I got to that point. Like, see, I got to the point where I can give you this whole guidance, and you're going to take this, you're going to split it apart, you're going to break through the different agents, you're going to coordinate it, and then they're going to be working, and then tomorrow I'll get to see it. In fact, this is probably a good example where we need to go back to increase the work throughout the night, because again, we don't have a lot of time and there's a lot of work to be done here, right? So we need to keep cracking on here now that I'm giving you all this feedback. And yeah, and that's fundamentally what we can do. So look, I think there's a lot of meat here. In fact, I think this is a good example of you probably want to do a couple rounds on this and get the storyteller to work, etc. Now, actually, what side effect on this? I noticed that there's a number of users, a number of agents that do one thing, stop and continue. Don't limit the agents by 5 or 10 or 20 minutes, right? You know, if you have agents firing up every hour, it's okay for them to sometimes take 45 minutes to do something, right? They should never take more than an hour, because that then becomes a bit weird, and that means a section fires before the other one. But it's okay to have, you know, long running, because you want to get rid of, you're going to do as much as you can for that one. In a way, you know, an agent that produces a lot of stuff in a session should clear his backlog, kind of, right? And then, so then if nobody assigns his stuff in the next session, he shouldn't have much to do, right? Okay. Wow, that was that was just a bit of page 9, right? So yeah, let me just finish off the page 9 bit. So wrong size, yet no things, and so the other thing here, right, is a lot of the times, sometimes you make time, or you mention somebody's time elements, like on page 9. Those are logistic stuff, but also if you need to do it, they are a footnote. In fact, we need to really explore the topic of footnotes, because we need to be adding footnotes, links all the time to the articles that we wrote, right? So we should be, you know, we have articles, we have newsroom.sk.ai, or, you know, other sk.ai websites. And they should be either a small link, a name in the bottom of the page, or a little QR code that we find. And that's just experiment. So that's what's the right size that your phone can still read it, right? But we should be sprinkling this entire book with all the references we wrote. Like, and this is the whole augmented reality thing. So maybe we need to figure out a way to connect this augmented reality, because it's like, hey, read more about this on this article. Like, we have some really powerful articles these days, like the fractal semantic graphs and how we think about this, and agents and stuff, and etc. So much stuff that's relevant to this. In fact, one of the things you need to then do is need to have a researcher probably do a pass at going again through the best articles that we have on the newsroom, best content, best vaults that we have, catalog that stuff, and then put it together, and then make sure that we link here, right? Again, that's the graph, right? And there should be a graph that, so you can even see the cross references between all of this stuff, right? --- *[Site index for agents](../llms.txt) · [HTML version](https://the-ultimate-insider.book.sgit.ai/brief/batch-04-page-9.html)*