WEBVTT 00:00:00.160 --> 00:00:12.880 For years, I've used a second brain. This might be the simplest, most powerful self learning personal knowledge base I've ever discovered built with Claude, and I'm gonna show you how to do it. Hi. At the start of 2026, 00:00:12.880 --> 00:00:15.545 one of the most respected voices in AI 00:00:15.545 --> 00:00:25.305 quietly posted how he runs his own personal knowledge base, a second brain where you hold all your information and make connections and use it to inform what you do. 105,000 00:00:25.305 --> 00:00:51.475 people bookmarked it, and probably almost none of them have built one. And that's the problem. This is genuinely the most useful AI setup I've seen in months and implemented in Claude, and it takes probably forty five minutes to build over a weekend. No Obsidian, no vector databases, no code, just a brilliant self improving knowledge base. Here's exactly what you're getting in this video. I'm gonna show you the architecture, 00:00:51.475 --> 00:01:08.680 the whole system in sixty seconds. I'm gonna show you the framework of how to build it and build it with you right now on this video, and then I'm gonna show you the Claude skill that helps you audit it and help it improve and maintain itself over time. The five step framework 00:01:08.680 --> 00:01:31.800 is this. You set it up. You dump your information into it. You then get AI to build a wiki. You ask it questions and create a compounding link to save answers back into it. And with a health check and that loop, it just keeps improving over time. So by the end of this video, you'll know exactly what the system is, why it beats every Obsidian Plus plug in setup you can find for simplicity 00:01:31.800 --> 00:01:38.680 and how to build your own right now with Claude. Now I found that day one of running it, your knowledge base is pretty basic. But day 100, 00:01:39.115 --> 00:01:42.955 it's a company asset that nobody else has. Your perspective, 00:01:43.115 --> 00:01:53.960 your sources, your judgment in one place. So double check you're actually subscribed to systems made better right now because YouTube might just be feeding you this anyway, and let's get on with all becoming significantly 00:01:53.960 --> 00:01:55.880 more intelligent very quickly. 00:01:59.880 --> 00:02:13.955 So here's the top level design in just sixty seconds before we go ahead and build it. Essentially, you're looking at three folders and one file on your computer that Claude looks at. I'm putting this right inside my Cowork OS, 00:02:14.115 --> 00:02:41.345 and I'm gonna be adding it to the templated system soon. You've got a Claude MD at the top of the knowledge base, which is the schemer. It directs Claude on how to read it and use it. You've then got three folders. Raw. Think of raw as your junk drawer. Articles, notes, screenshots, meetings. You just everything goes in here when you save it, you don't organize it. Then you've got the wiki where AI writes the organized version. You never edit this by hand. It's all done by the AI. And then you've got outputs, 00:02:41.505 --> 00:03:03.950 answers, briefings, and reports that the AI generates when you ask it questions. Now the best bit is those then get fed back in and help to refine it, plus one file at the root. Yeah? The Claude MD. And you could have multiple versions of this within essentially a top level folder where it all sits. That means you can have multiple knowledge bases all connected together. That's it. No database, no Obsidian, 00:03:03.950 --> 00:03:18.250 no vault setup, just folders and text files on your computer. And before you ask, no. You don't need a rag embedding or any vector store if you know what all that stuff is. Kapathi's own knowledge base is around a 100 articles and 400,000 00:03:18.250 --> 00:03:56.530 words, and the LLM handles it fine maintaining an index and reading what it needs. If it works for one of the most respected AI researchers alive, it'll probably work for your business. The best thing, like what I've done in my Notion Agent OS, is you can then point a custom agent at that knowledge base, and it becomes a specialist agent expert that you can speak with. It can use the knowledge to work on problems with you. But that is for another video on the channel. I'll be sharing a video soon about how I'm turning bodies of work from expert thinkers into personal assistants that help me on my business. It's totally wild. And we're doing that in both Notion and Claude. 00:03:59.570 --> 00:04:23.030 Okay. I am doing this in Claude Cowork. We've got a new window open, and I'm pointing it at my main Cowork OS folder. So basically, I have everything in one folder on my home, the local. There's a Claude Cowork folder. Everything happens in here. You just direct at it, and I've got instructions like about me files and all of that. But watch my how to get set up on Cowork first if you wanna do that. But we're gonna add a new folder in here called 00:04:23.190 --> 00:04:23.990 knowledge. 00:04:23.990 --> 00:04:40.365 There it is. Gonna drop it in at the top level. We're gonna go back to Claude, and we're gonna set this up. So I use WhisperFlow to instruct Claude on what I wanna build. Link in the description. This is what we're gonna say. I want to build a self improving knowledge base that you manage as a librarian. 00:04:40.365 --> 00:04:50.620 Let's start and make a folder structure inside the new folder I've added in your Claude Cowork folder called knowledge. Inside that, I want three subfolders. 00:04:50.780 --> 00:04:52.700 We want raw, wiki, 00:04:52.940 --> 00:04:54.060 and outputs, 00:04:54.140 --> 00:04:57.655 plus drop a Claude MD file in the root, 00:04:57.895 --> 00:05:06.135 and I'll show you what is gonna go in that in a moment. And what you could do is give it context of what you're doing. So we could say, for context, 00:05:06.375 --> 00:05:09.255 here is Andre Kapathi's 00:05:09.255 --> 00:05:31.085 explanation of what we're about to build, but we are gonna be doing this locally rather than with Obsidian. I'm gonna use Opus 4.7 because it's intelligent and it will do the work, but probably don't need it. And there it is. It's turned up. Let's drop those in. I'm going to just rename this so it's clearer. We'll call it knowledge base. So interesting it couldn't read the Twitter thread, but I'll just put it in here. 00:05:31.485 --> 00:05:33.005 Here's the information 00:05:33.085 --> 00:05:34.525 for you from that 00:05:35.005 --> 00:05:38.045 thread. However, I'm gonna take you through step by step what I want. 00:05:39.590 --> 00:05:48.230 Now if we go back and take a look at the folder, we've got our knowledge base. Now what we might wanna do at our top level is create another one, secondary knowledge, 00:05:48.230 --> 00:05:50.630 and we could drop the whole thing inside 00:05:50.630 --> 00:05:57.005 that, And we could give it a subject. So what do we want this to be on? Why don't we make this one on productivity? 00:05:57.165 --> 00:06:03.325 Alright. I've dropped what we just built into another top level folder, which is called second brain 00:06:03.325 --> 00:06:09.430 knowledge. And that's gonna be the top level where we can create multiple versions of knowledge bases. 00:06:09.510 --> 00:06:15.350 So based on this information and what we've created, please create me another Claude MD file for that folder, 00:06:15.510 --> 00:06:17.190 which will explain 00:06:17.190 --> 00:06:19.145 the basic layout 00:06:19.145 --> 00:06:30.985 when a new knowledge base is created in its folder. Second, I've renamed the knowledge folder to be a productivity knowledge base, which is what we're going to do. And here is a basic template of what I think 00:06:31.420 --> 00:06:54.215 the Claude MD file for each knowledge base should look like, but please make suggestions and a plan how we could make this really strong and improve on it. And then what I've got is a little example of what I think it might look like. So something like this, how it's organized, what it does. So we're gonna drop that in. Okay. Great. And you can work with Claude to improve this. So giving it that Kapathi example, 00:06:55.015 --> 00:07:01.575 it said these are things that we're gonna need in your Claude MD. So I'm gonna add these in. We wanna make it standardized. 00:07:02.150 --> 00:07:13.270 We wanna know how health checks work and when and how to ingest stuff. It's got a plan. Okay. Great. I think, ultimately, we will set this to be active as a librarian 00:07:13.590 --> 00:07:23.045 or between active and aggressive. I think I would do this using scheduled tasks, and we'll set those up in a bit. But first of all, I think the main job is to 00:07:23.285 --> 00:07:32.340 write the basic Claude MD file for how this is gonna work with your best suggestions to keep it clean and simple, but powerful and effective. 00:07:32.740 --> 00:07:34.900 In terms of ingesting material, 00:07:35.060 --> 00:08:08.010 this will just be me doing this manually, but it wouldn't be unhelpful for us to add the option for you to work with me and guide me through it in a process. So you could work that in to the top level Claude MD. But I would like to, on the first pass of this knowledge base, input a load of stuff, and then you would build the wiki from there. So let's just create the first instructions. And as for monthly health checks, we'll come to that later in detail, but this is a basic suggestion of how this might work. And I'm gonna paste what I've written in, review the entire wiki directory, flag contradictions between articles, find topics mentioned, 00:08:08.250 --> 00:08:13.055 list claims not backed by a source, etcetera. Please write your proposed 00:08:13.455 --> 00:08:25.870 top level second brain MD and then knowledge base MD for the productivity knowledge base example we're building. So what I'm essentially asking you to do is based on this feedback is build me its best version, 00:08:26.350 --> 00:08:29.230 and it's informed by that Kapathi article 00:08:30.030 --> 00:08:39.895 that we showed it earlier, which was here. So it's kind of gonna follow this process, and it's now building what we need. So in the second brain knowledge base, 00:08:40.215 --> 00:08:43.655 we have a top level one. It's a container for multiple knowledge bases. 00:08:44.055 --> 00:08:48.535 And when it creates a new one, we ask it to do that. And this is how the system works 00:08:48.900 --> 00:08:51.140 and how they are independent. 00:08:51.300 --> 00:09:03.220 Nice. And then the detailed behavior is for each system. That's great. Then it should be working on one in here, and now it's building this for us. So it suggested that the top level file would have a guided ingestion 00:09:03.220 --> 00:09:11.125 mode to call on. You can see what it's up to here. Okay. It's done it. And here we go. Now what I do wanna make sure I've done in my actual Claude MD, 00:09:11.285 --> 00:09:13.685 we open this up, the focused areas. 00:09:13.765 --> 00:09:28.770 So list three specific themes this knowledge base will deepen. I'm gonna change that. This knowledge base is focused on the ethos of doing less but better, finding a balanced approach to a greater contribution to the world, 00:09:29.250 --> 00:09:30.290 deeper thinking, 00:09:30.690 --> 00:09:39.515 and stronger output whilst managing health, happiness, and balance in your life. So the themes are attention and energy management, 00:09:39.675 --> 00:09:42.075 systems design, deep work, 00:09:42.315 --> 00:09:43.515 essentialism, 00:09:43.675 --> 00:09:45.275 and effective 00:09:45.435 --> 00:09:46.555 contributions 00:09:47.460 --> 00:09:48.900 through productivity 00:09:49.700 --> 00:09:50.740 principles. 00:09:50.820 --> 00:09:56.180 Now one question I have is whether we need a memory file that simply lists 00:09:56.180 --> 00:09:56.820 when 00:09:57.380 --> 00:10:00.180 the last action was taken so that 00:10:00.615 --> 00:10:06.775 the process that's automated knows what is new in the raw files and what is 00:10:07.175 --> 00:10:08.535 already processed. 00:10:09.735 --> 00:10:22.210 Good. So it agrees that we need a memory to make sure that it knows when it last processed something, and we can add that in. This is great. We're all set. Now, of course, you can do all of this manually, but I really like the idea of this being quite automated. 00:10:22.290 --> 00:10:51.830 Great. It will make that on the first pass. Excellent. Now, of course, I'm building this as I go. I'm learning it as I go, and I will share my final template version for this linked below if you want to try it. But it will be part of Cowork OS, so check that out after this. So step one is essentially that build that system out. And so what we should now have is a knowledge base with the MD ready to go, which will explain how everything works. It it talks us through the process, the folder structure, 00:10:52.230 --> 00:11:05.055 and what the change log MD will be. Doubling as a systems memory, it talks it through how to do things. You don't need to worry too much about that right now. Uh, that is the plan, and you can ask Claude to do it for you. We've then got our outputs, 00:11:05.215 --> 00:11:19.940 which will be things that it creates for me, raw and wiki. So next up, we need to do step two, which is the dump. And I think this probably for most people might take like ten minutes just to find everything they currently have and put it into the raw folder. Pretty simple. I'm just gonna do that quite quickly. 00:11:23.555 --> 00:11:37.590 The issue many people miss about using a second brain first. Now if you've spent any time on Twitter, x, you've watched the same cycle play out a 100 times. People post a screenshot of their Obsidian Vault or Notion setup, 00:11:37.750 --> 00:12:01.305 linked notes everywhere, graph views, plug ins, people bookmark it, and then you kind of forget about it. And to be honest, I've tried this myself. I've built these in Notion on my computer and everywhere. This is the simplest way. But this is the point about a second brain as well. We find something brilliant, we save it, and then we lose it. The fix is a second brain that actually works intelligently for you. Okay. Great. So next, I want to 00:12:02.340 --> 00:12:10.020 ingest and dump all of my current knowledge on productivity into our first trial knowledge base, the productivity and knowledge base. 00:12:10.340 --> 00:12:13.060 To do this, why don't you find 10 to 20 00:12:13.140 --> 00:12:21.645 strong entries in my knowledge base in Notion that can be found here. So what I'm gonna do is jump over to Notion, go into my knowledge and research, 00:12:21.805 --> 00:12:47.960 and we've got a bunch of stuff in here. So I'm just gonna give it the link to this database. So why don't we actually, like, view the entire database and get the link to it? We'll go back in, paste that there. I may also attach couple of files here for you. So you can, of course, also click this and just add files or entire folders, whatever you wanna do. But we're just gonna try this as a little example. And while that happens, I'll show you how that's working. In customize in Claude Cowork, we can go to connectors, 00:12:48.565 --> 00:12:50.805 and I've connected up Notion, 00:12:50.805 --> 00:13:20.525 so it means that it can now go and action find and draw stuff from it. So this is just an example, but it's also worth saying that in my system, have an about me section and a context map. And that context map shows all of the key databases in Notion which it can read from. So in many ways, it should have already known that. I didn't actually have to show it, but I really like that approach to have a context map. I think try to find good examples of longer form entries and clippings, articles, 00:13:20.605 --> 00:13:44.555 or quotes from books that have been added rather than the AI research sets. Now while it does that, I wanna say something to you. You don't need to be tidy when you do this. Just copy and paste articles, notes, screenshots, meetings, transcripts into raw. You can even just paste them into the chat and get the AI to add them for you. Don't make this pretty. The point is it's a folder for capture. The organization 00:13:44.555 --> 00:13:51.355 is the AI's job, and that is why this is so nice to do. For example, here's a blog from Cal Newport on deep working. 00:13:51.515 --> 00:14:06.750 And what I might do is just literally take all of that, copy it, and paste it in here. Please add this from Cal Newport's deep work post. PS, when we add stuff to the raw file, you just add this as an MD file. Images can also be attached into it from me. 00:14:07.230 --> 00:14:14.510 For example, I've got a my PDF here of how to build a agentic business. I'm just gonna take that and drop it into raw. 00:14:15.285 --> 00:14:47.805 PDFs are probably harder to read. I think the AI has more trouble with that. I'm gonna put it in as a test. Great. So it's fetching a bunch of stuff from Notion as an example here. Now one little tip though, if you're doing this manually, you can use Xcode. It's a free Mac desktop app. In there, you can create markdown files. So I've just pasted an article into one from Gretchen Rubin here and added it into the folder. So really quickly, you've added something in. So if you wanna do that, you just are going to open Xcode. You're gonna do file new 00:14:47.805 --> 00:14:50.365 from template, and you just wanna find 00:14:51.565 --> 00:14:55.085 markdown file. You just select that, create a new file, 00:14:55.405 --> 00:15:17.265 and you can name it, drop it in, and you're you're good to go, basically. That's how it would work. So that'd be a really quick way to manually add markdown files in. But for a lot of people, um, you'd probably be able to just share the information with Claude and get it put in. That does cost credits, though. So it's up to you how you wanna do it. If you do choose to use Obsidian, they have a great web clip clipper browser extension 00:15:17.425 --> 00:15:28.620 that converts any page into a clean markdown file in one click, and that's free. So it's worth checking out. So here we go. We've got a bunch of things in here. We've added them all as markdown files. We've got the 00:15:29.820 --> 00:15:34.060 PDF that I added. We've also got the one I added using Xcode, 00:15:34.060 --> 00:15:35.340 similar situation. 00:15:35.580 --> 00:15:50.275 And you can, of course, also just go in to your downloads folder and just drop images in. So there's a JPEG there, which is a nice example of the process that we're working through by Corey Gammon. Cool. So we've got our raw input. 00:15:50.950 --> 00:15:53.910 My Claude system also created an ingested 00:15:53.910 --> 00:15:55.030 registry, 00:15:55.190 --> 00:16:07.705 so it talks about when everything went in, which is useful. Okay. Step three is build the wiki. This probably would take you around thirty minutes. You're gonna point Claude at the folder and give it one prompt. 00:16:11.305 --> 00:16:18.025 Read everything in raw and compile a wiki in the wiki folder following the rules in your Claude MD. 00:16:18.650 --> 00:16:31.050 Create the index MD first, then one MD file per major topic and link related topics. And then you basically walk away and let it do the job. What you come back to is information 00:16:31.050 --> 00:16:31.610 organized. 00:16:31.935 --> 00:16:36.015 Topic pages with summaries, connections between ideas you didn't know existed, 00:16:36.175 --> 00:17:02.925 an index that makes everything searchable in second. Now, the problem with something like Notion or Obsidian to manage a second brain for knowledge like this is that they kind of ask you to be the librarian. You organize things yourself. You make the links. You manage the tags and folders. You configure plug ins, all the rest of it, and then it kind of goes by the wayside. What I think Kapathi has figured out with this approach using LLMs is that the AI becomes the librarian. 00:17:02.925 --> 00:17:27.515 You dump information in, Claude organizes and links it, summarizes it, and indexes it. And by the end, it's learning and improving on its own, helping you actually apply the knowledge to output. Think what this could do for your team, your business, or just your personal output as someone exploring ideas and work. Okay. So it's working through. You'll see it's created a index. It's written foundational 00:17:27.515 --> 00:17:28.555 articles, 00:17:28.635 --> 00:17:31.515 and then it's gonna do method articles, thematic articles, 00:17:31.675 --> 00:17:39.035 and then write a questions MD and a change log. Now one little tip when you get your AI to do this is to make sure that it's read 00:17:39.320 --> 00:17:41.080 your anti AI 00:17:41.160 --> 00:17:41.880 writing 00:17:42.120 --> 00:17:43.080 style guide. 00:17:43.160 --> 00:17:45.960 Now what that actually looks like in my world 00:17:46.360 --> 00:18:11.370 is a very similar process to what I've done in my Cowork OS template. I have this template in there, and it's really a writing rules MD. And this is built on the Wikipedia anti AI writing style. So if you look up AI writing style on Wikipedia, paste that into Claude and say, create yourself instructions to never do any of this. It just avoids bad writing, essentially. I'm not gonna go into it much further than that. But I've made sure that 00:18:11.450 --> 00:18:19.290 Claude, as it's writing its wiki, which you can see it's starting to happen here. Look. We're getting all these different things. It's doing that using 00:18:19.290 --> 00:18:27.695 the writing style guide. It's also great to see here, if we just take a little look, there's loads of information going in, but it takes up so little 00:18:27.855 --> 00:18:28.735 storage. 00:18:29.055 --> 00:18:30.975 Four KB. It's nothing. 00:18:31.215 --> 00:18:42.820 Uh, and this is the joy of MD files. So let's see what it's got to. Now I suspect you may be aware that this process is quite demanding. In order to pull this off, you are gonna probably either need to do it in sessions, 00:18:43.140 --> 00:18:53.675 or you're gonna want to be included on a max plan like I am. So if we go and look in settings, let's take a little look. I've been doing other things on here, but under usage, we're 39% 00:18:53.675 --> 00:19:00.715 into my current session. And actually, news. Claude recently announced that they are doubling usage limits 00:19:00.715 --> 00:19:02.155 across sessions 00:19:02.590 --> 00:19:22.585 and during peak hours. That's not weekly limits, but it is session limits. And you can, of course, turn on extra usage, but I don't recommend it. I just got a free spend, which is nice. Okay. So we have now built our first knowledge base. We've got our top level knowledge base here. We've got a Claude MD that instructs 00:19:22.585 --> 00:19:23.945 us how to build 00:19:24.345 --> 00:19:25.705 knowledge bases 00:19:25.865 --> 00:19:27.065 and their structure 00:19:27.145 --> 00:19:51.175 and what they look like, which means that this can be a a global knowledge base with lots of individual ones, and we've got that. I've actually got a little memory file here, which shows us that I have a place where I keep my projects that I'm working on. And this is really just a memory and a project beef brief for building a project knowledge base, so you don't need to worry about that. And then this is what you've built. You've built a project knowledge base. It has a change log with 00:19:51.655 --> 00:19:57.495 the most recent entries when things have happened, and it has the main Claude MD that instructs 00:19:57.575 --> 00:20:18.125 the system how to work. So when you do this, make sure you download the templates to get you started on the process from the link below. Then we have our raw. So I've got a bunch of example raw entries. They're all things that have just gone in like this. And then we have our wiki, which is all of the things it's created. So it created an index, which shows the key concepts 00:20:18.365 --> 00:20:23.805 within the system so far. And then within that, we then have all of the individual entries 00:20:24.045 --> 00:20:29.690 for, like, specific subjects. So effortless state, energy management, 00:20:30.170 --> 00:20:33.210 habit formation. So you you see these become themes, 00:20:33.290 --> 00:20:34.170 frameworks, 00:20:34.250 --> 00:20:36.970 templated ideas that are directed by the system. 00:20:40.955 --> 00:20:58.570 So now we need to ask it questions and get things out of it, and it's this process that actually changes everything. Because every time you ask the agent a question you like the answer to, you can then save that back into raw or into the wiki, and the system gets smarter the more you use it. So each question 00:20:58.650 --> 00:21:01.130 makes the next answer better, 00:21:01.370 --> 00:21:11.450 and that is because it's gone into outputs. So what we're gonna do is just test this first of all. So I'm gonna start a new window. I'd like to test out my new productivity knowledge base, 00:21:11.925 --> 00:21:14.085 and I have a question to ask you 00:21:14.325 --> 00:21:16.565 based on the knowledge base. 00:21:16.725 --> 00:21:24.245 What's the best way for me to balance achieving a huge amount in a short amount of time whilst managing my energy, 00:21:24.325 --> 00:21:35.980 happiness, and health? So it's found the productivity knowledge base. It's reading the index. That's promising. Reading the most relevant wiki entries. This is a test. It should end up in here. It's comparing 00:21:36.380 --> 00:21:37.260 Newport, 00:21:37.420 --> 00:21:42.645 McEwen, and Berkman and Forte all covering this answer. You can't win both simultaneously. 00:21:42.645 --> 00:21:48.245 Trying to is what produces burnout. So you can't do loads of work and rest. 00:21:48.565 --> 00:22:06.480 So seven things the knowledge base says about that I can actually do. This is really cool. I really like this, and it's referencing where stuff is coming from. The test went well. The wiki had the article in every angle of your question. This is all great. Okay. So we now need to check if it actually did an output. Let's have a look. Well, it didn't. So okay. This is great. 00:22:06.720 --> 00:22:09.840 But we should have a rule within this system, which is when I ask a question, 00:22:10.465 --> 00:22:17.905 the report is generated into outputs so that we are gaining deeper insights. 00:22:18.385 --> 00:22:19.905 So please, a, 00:22:20.305 --> 00:22:26.340 update the Claude MD to ensure that this is always the case. B, turn this into 00:22:26.500 --> 00:22:28.500 a report that goes into outputs. 00:22:28.660 --> 00:22:29.620 And c, 00:22:29.700 --> 00:22:32.260 then rerun the process with this query. 00:22:32.500 --> 00:22:50.455 Based on everything in the wiki, what are the three biggest gaps in my understanding of this topic? If we go back in here, please make sure you first reread the Claude MD for the knowledge base as I've now updated the topic focus. And I'm also gonna add one more thing in here, which is write me a 500 word briefing 00:22:50.630 --> 00:23:01.430 on doing less but better using only what's in the knowledge base. Great. So I'm gonna ask that. So we're doing a couple of things here. First of all, we're refining the system to make sure that it always 00:23:02.150 --> 00:23:03.990 generates reports into outputs. 00:23:04.725 --> 00:23:05.525 Secondly, 00:23:05.845 --> 00:23:11.525 we want to turn the report it's just created into outputs. And thirdly, I'm gonna give it two further tests, 00:23:11.765 --> 00:23:21.290 uh, answering these two questions. And I'm asking it to make sure it rereads the Claude MD for the knowledge base, and now I've updated the topic focus. So if we go and take a little look at the 00:23:21.690 --> 00:23:28.010 outputs and look at what it's written for us, we can see some great results here. It's saying it's looked at all the articles, 00:23:28.170 --> 00:23:33.045 and it said it has almost nothing on the journey from where most people start, overcommitted, 00:23:33.045 --> 00:23:35.525 fragmented attention, default on connectivity. 00:23:35.605 --> 00:23:36.245 Cool. 00:23:36.405 --> 00:23:37.525 It's missing 00:23:37.765 --> 00:23:41.285 the mechanics of stopping, and then it's got real decision 00:23:41.605 --> 00:23:48.370 method for what counts as essential. That's missing. It's missing working with other people, interestingly. It generally presumes 00:23:48.370 --> 00:24:23.170 that you're working on your own. This is really cool. So what we could now do is go and use this to feed into the updates and improvements, and we can actually get the AI to build and improve on itself. Make sure that your instructions say read the outputs and work from there. So now I'm gonna show you step five, which is the health check, and this one really matters. The AI will sometimes write something slightly wrong, you'll save it back, and the next answer quietly builds on a mistake. So once a month, you wanna audit this. And the prompt is gonna be something like this. 00:24:26.370 --> 00:24:41.045 Now I'm gonna show you in a moment how to build a scheduled task and the skill to do that. But first, let's just do this really simply. And to do it, I'm actually just going to point this directly at the folder to demo this. We're going to co work. We'll go into 00:24:41.800 --> 00:24:42.840 knowledge base 00:24:43.640 --> 00:24:44.920 and this folder. 00:24:46.120 --> 00:24:53.000 So if you just do this manually, you wanna say something like this. Please review the entire productivity knowledge base wiki. 00:24:53.385 --> 00:24:54.745 Flag contradictions 00:24:54.745 --> 00:24:57.305 and inconsistent data between articles, 00:24:57.465 --> 00:25:00.745 find missing data, and fill the gaps with web search, 00:25:00.905 --> 00:25:04.505 list claims not backed by a source in raw, 00:25:04.825 --> 00:25:06.265 and suggest connections 00:25:06.265 --> 00:25:10.890 between articles I haven't drawn yet and three new article candidates. 00:25:10.970 --> 00:25:12.250 So this is quality 00:25:12.330 --> 00:25:12.890 control. 00:25:13.210 --> 00:25:40.930 The one thing I am gonna write here though is, please do not invoke my health check skill. This is a demo of just doing it clean with this instruction because I've created a health check skill. Let's try it. And for now, as this is a demo, actually, please just share your results and changes in the chat. Don't edit anything currently in the wikis. I'm just gonna say that as well because I want to you just see the kind of thing it's gonna do. So what you can see it's now doing is reading through the wiki and the system 00:25:41.250 --> 00:25:53.985 and making a complete audit of the knowledge base. Now you would just set this going, leave it, and come back. But even better, we can schedule it. And while it does that, I'll show you what that scheduled task looks like. If we go into scheduled, 00:25:54.465 --> 00:25:56.945 we now have this knowledge base monthly 00:25:56.945 --> 00:25:59.745 health check. All I did here was ask 00:25:59.825 --> 00:26:01.745 the system to create me 00:26:02.065 --> 00:26:07.240 a automated health check comprising of a knowledge based 00:26:07.320 --> 00:26:11.080 health check skill that it would create with its skill creator 00:26:11.240 --> 00:26:22.435 plug in. And it basically says, go through and do the things that we've just asked for based on the skill. You can set this up so that it runs on different times, but interestingly, 00:26:22.755 --> 00:26:29.715 you can have a custom schedule. So if you ask it when you speak in the chat to build you a skill that is monthly, 00:26:29.795 --> 00:26:38.570 it can do that, uh, not just follow the options that are in the selectors. And that's it, basically. It's ready to go. And you can get it to act without 00:26:38.730 --> 00:26:45.530 pausing for approval if you want. That's an option. So I'm gonna save that for now. And then if we go into customize, 00:26:45.530 --> 00:26:53.485 I've also created in skills this knowledge base health check skill, and this will work its way through the process. 00:26:53.885 --> 00:26:57.565 And it does it in two phases. It has a first audit and file process 00:26:57.645 --> 00:27:03.485 where it reads my writing rules guide for anything that it's gonna write. It reads the change log, the wiki, 00:27:03.645 --> 00:27:04.765 and what's been ingested, 00:27:05.310 --> 00:27:09.550 as well as the outputs that have been created since the last last health check. 00:27:09.870 --> 00:27:15.390 And then it runs a seven stage audit. And the seven stages are these, contradictions, 00:27:15.710 --> 00:27:17.950 broken backlinks and orphaned references, 00:27:18.190 --> 00:27:19.310 source provenance, 00:27:19.835 --> 00:27:20.635 coverage 00:27:20.635 --> 00:27:22.555 that the raw files have, 00:27:23.115 --> 00:27:28.395 stale articles, anything that's out of date, older than ninety days and not relevant, and suggested new articles. 00:27:28.555 --> 00:28:02.940 And then this is a report template of how it gives a report. And then has a second phase, which is if you're doing this interactively, if you're actually directly asking for it, it will also ask which findings to action in ask user question. So it means you can kinda go through it fully and then fully, uh, commit it. In the phase one, it will just give us a report that we can then ask to be actioned later on. Now I'm creating a template version of this for you guys so you can just download it via the link in the description and use it. But for now, let's go and see what our example is up to. And here is our audit. Let's see what it says. Effort versus effortlessness 00:28:02.940 --> 00:28:04.300 are contradictions, 00:28:04.380 --> 00:28:05.180 inconsistent 00:28:05.180 --> 00:28:06.620 numbers and framing. 00:28:06.700 --> 00:28:09.020 Nice. It's cleaning up attribution 00:28:09.020 --> 00:28:09.740 drift, 00:28:09.820 --> 00:28:12.220 unsourced and undersourced claims, 00:28:12.700 --> 00:28:15.580 building a second brain, mood first productivity 00:28:15.580 --> 00:28:17.345 has not captured the link, 00:28:17.585 --> 00:28:25.345 habit formation, so on and so forth. Gaps the wiki has. There's no underlying research for the cathedral effect. We don't have the book. 00:28:25.505 --> 00:28:26.465 An unprocessed 00:28:26.465 --> 00:28:34.820 file that we haven't ingested. Great. That's something I added recently, an unaccounted JPEG, and then it's found some really interesting connections that we might 00:28:35.060 --> 00:28:36.260 not have seen. 00:28:36.420 --> 00:28:43.220 So quick verdict. It's unusually clean for an early stage knowledge base. All looks pretty solid. Main weaknesses, 00:28:43.300 --> 00:28:44.180 attribution, 00:28:44.180 --> 00:28:46.775 unprocessed raw files, not naming 00:28:46.775 --> 00:28:48.055 the underlying study, 00:28:48.295 --> 00:28:49.975 philosophical contradictions. 00:28:49.975 --> 00:28:55.095 Okay. Great. We'll leave this here as I'm now gonna start a new session and compare this with my skill 00:28:55.335 --> 00:28:56.695 and triggered 00:28:57.095 --> 00:29:05.150 scheduled task to see how the results compare. So now we're gonna start again, and let's run my scheduled task. So we can actually go to scheduled, 00:29:05.630 --> 00:29:08.350 click into the scheduled task, and click run now. 00:29:08.590 --> 00:29:10.830 As simple as that. Now if we go into 00:29:11.765 --> 00:29:15.285 the knowledge base so you can see it's now implemented 00:29:15.285 --> 00:29:17.605 the knowledge base health check skill. 00:29:17.925 --> 00:29:25.500 It's following that now and reading my writing rules. These are anti AI writing rules. We can see that it's gonna have checked the latest 00:29:25.980 --> 00:29:27.820 item in the change log. 00:29:28.460 --> 00:29:29.580 So these are the 00:29:29.980 --> 00:29:31.340 latest updates. 00:29:31.340 --> 00:29:32.940 It's working chronologically. 00:29:32.940 --> 00:29:44.335 It's then gonna read through all of the other files, and we should see it now work. So let's let that run and see what we get back. Oh, and if you're interested in this item here, push summary to BriefBuddy, 00:29:44.335 --> 00:29:48.975 I've actually created myself a little reporting app that is automatically updated 00:29:48.975 --> 00:30:36.460 and turns up on my phone. You don't need to do this. Uh, the system will just essentially, uh, you'll see when the scheduled task has run, you'll see a little, um, blue dot for something, and you can go and look at that and find the report and the brief. So this, for example, is another scheduled task that I'm running, um, essentially draft stuff so I can go and look at them and work on it. So when this goes blue, we'll be ready to see what's happened. Okay. Great. So that took it about twelve minutes. Now it is worth remembering that this probably is gonna cost a few credits to do it. That's why I'm only scheduling this to be monthly, and you might wanna do it for each knowledge base you build on a different day. So you don't just use all your credits up, but it's a really useful thing to be doing to make it powerful. So we can see it turned up. You don't really need anything more than that. If you come into Claude, you'll see that this has happened, and we can click on it in either position. 00:30:36.620 --> 00:30:51.195 We can go in and take a look, and we can see it's completed it. It's run that. It's filed a report. The brief buddy thing, I'll need to problem solve. But to be honest with you, I don't really need it to do that. And it will show them to us here, but we can just go over to our folders to see what's happened. 00:30:51.515 --> 00:31:08.130 The change log, first of all, will have been updated today. There you go. Health check first run, and it's reported on what's happened. So the system will know where it's at. That's great. And then in outputs, we can see here is our health check. There you go. We've got the wiki is unusually well aligned. It's done a similar thing. 00:31:08.845 --> 00:31:13.405 New candidates, so it's gone through and looked at the issues and discoveries. 00:31:13.405 --> 00:31:14.765 No stale articles. 00:31:15.485 --> 00:31:19.085 It's cleaned up some banned words, American spelling, 00:31:19.245 --> 00:31:26.850 and then we've got suggested new articles. This is probably where the real value is. So it's suggesting we look at collaborative productivity, 00:31:27.170 --> 00:31:28.690 good habit recipes, 00:31:29.330 --> 00:31:32.050 looking at b BJ Fogg, interesting. 00:31:32.370 --> 00:31:34.930 And we've got effort versus effortlessness, 00:31:35.170 --> 00:31:37.970 making the frame easy, accepting strain inside, 00:31:38.050 --> 00:31:50.915 interesting. And then it's got an action menu. So for phase two, things that it could run. And we could now ask it to run those things, and we will get that automatic update. So this is a reasonably, like, in-depth process. 00:31:51.155 --> 00:32:04.820 You could always simplify it. It really comes down to what you wanna do, but check out the templated options in the description, and you can take it from there. Or if you're downloading my coworker OS, it will be baked in. So as a final example here, I'm gonna get it to actually update. 00:32:05.735 --> 00:32:17.415 Please see the latest health check-in your productivity knowledge base and run the action list from it on that knowledge base. Now what you can see is it's it's written itself a great list. It's applying the writing rule fixes. 00:32:17.655 --> 00:32:22.180 It's adding the new stuff to ingest, drafting the new articles, 00:32:22.180 --> 00:32:29.060 and then it's gonna update everything, which is great. It's worth saying, I think for most people, once you've tested it in these two stages, 00:32:29.460 --> 00:32:38.075 it's quite easy for you just to have it automatically do it. So you could just say, just do the work, Report an action. I think that's better. And, potentially, 00:32:38.235 --> 00:32:44.075 you kind of refine your instructions to make it rigorous but not cost you loads and loads of credits. 00:32:44.315 --> 00:32:45.515 As an example, 00:32:45.835 --> 00:32:51.580 for the example that we've run, so the first one I did without without the skills, then the one with the skill and this, 00:32:51.820 --> 00:32:52.700 the usage 00:32:52.780 --> 00:32:56.220 on my max plan for this current session is at 45%. 00:32:56.380 --> 00:33:14.885 That's on a five x max plan. So that's significant use of credits. But once a month for a really powerful knowledge base, not too bad. Let me know in the comments how you feel about that. And here we go. We've got the results. It's created new articles on habit receipts, working with others, and effort versus effortlessness 00:33:15.400 --> 00:33:18.280 into the gaps that we're missing. It's updated 00:33:18.520 --> 00:33:22.600 the index questions and change logs, and they're all ingested, 00:33:22.600 --> 00:33:34.515 which is really cool. It's giving me a bit of feedback about some web search stuff, and then it's created the documents. And if we go and check out the files, we'll see the new items have been ingested. 00:33:34.675 --> 00:33:39.395 And the new entries in the wiki have been added, which is great. 00:33:42.850 --> 00:33:53.570 So it should be that we now get a very different result. So if we ask in a new task, take a look at the productivity knowledge base and give me a report on how I can balance 00:33:54.055 --> 00:33:54.935 making 00:33:55.175 --> 00:33:57.335 serious and useful effort 00:33:57.415 --> 00:34:13.710 versus making my week and days feel effortless in how I contribute to my life. These are just examples. Right? But let's just drop it in and see what it gives us. And it's created it. Now annoyingly, it's not presented it to me. Please, can you update your Claude MD files and the templated one for 00:34:14.110 --> 00:34:15.310 knowledge bases 00:34:15.870 --> 00:34:53.535 so that any report that's created in response to a question is presented as a clickable page to open in the chat. Great. There we go. And it's now showing it to me, so I can actually click on it and read it. And this is what it's given us, a little report. Now here's a nice little tip. If you ever wanna make your learning easier when you're doing this, I use Speechify to read things back to me. So I can just do control option a, and it reads it. The question, how can I balance making serious and useful effort versus making my week and days feel effortless in how I contribute to my life? Those are the five steps to building, refining, and using a knowledge base that learns as it goes. 00:34:56.575 --> 00:35:07.250 So here's the bit you need to remember to take away with you. Day one of running this, your knowledge base isn't gonna do loads. It's got whatever you dumped in over the weekend, useful but not revolutionary. 00:35:07.410 --> 00:35:08.690 But day 100, 00:35:08.690 --> 00:35:20.685 then you've actually built something valuable. Every meeting transcript that mattered, every answer you've saved back into the system becomes a carefully curated cross reference linked and summarized set of information 00:35:20.765 --> 00:35:23.565 that you can query with the librarian 00:35:23.565 --> 00:35:24.445 themselves. 00:35:24.685 --> 00:35:43.195 And it's that kind of asset that's nearly impossible to replicate because nobody else has read what you've read or saved what you've saved. So if you only do one thing from any video I make this year, do this. It's forty five minutes on a Saturday morning, and you'll thank yourself in three months. One last thing, everything we built today, 00:35:43.355 --> 00:36:14.905 the folders, the Claude MD, the prompt, the health check, they all ship inside the final version of my Claude Cowork OS when it comes out. It's in beta as I film this and you can download it right now. It's been brilliantly received so far. The whole point of it is to help you skip past the fiddly setup bits and land on a working Claude environment faster than most people manage on their own. It's been a game changer for me and a lot of others using it. And, of course, you can watch the video that shows you exactly how to do all of that right here. I'll see you on the next one. Bye.