00:00:00If Claude code plus notebook LM is amazing and Claude code plus obsidian is
00:00:04free value and Claude code plus the brand new skill creator is legitimately game
00:00:09changing. Then what's going to happen when we combine all these tools together in a
00:00:13practical yet simple to set up workflow that you can start using today and under
00:00:1930 minutes. Well, that is exactly what we are going to find out in today's video is
00:00:23I show you step-by-step how to create one of the most powerful workflows inside of
00:00:28Claude code. This workflow turns Claude code into an absolute research monster.
00:00:33And this video is also pretty much a capstone of everything we've talked about in the
00:00:37last few videos, because we've covered things when it comes to Claude code and
00:00:40notebook LM and Claude code and obsidian and Claude code and the new skills
00:00:43creator. But here's where we take all these lessons and we synthesize it into
00:00:47something that has practical value. And on that note, what's important isn't my
00:00:52exact use case, right? This is a personal chase AI use case, right? And how I do
00:00:57research for my content, but you're not a content creator. You probably have a real
00:01:01job. So what I want you to focus on throughout this entire lesson, isn't the
00:01:05exact intricacies of how I'm doing my YouTube search. You should be focused on how
00:01:10do I swap the YouTube search for whatever use case I have and whatever source of
00:01:14information I need, whether that's PDFs or articles or texts or whatever, right?
00:01:18How can we fit in this template into your life? That's where the value lies. And
00:01:22that's what I want you to focus on. And it's also something this is great at,
00:01:26right? This is a very flexible workflow that can adapt to your needs. And we love
00:01:32that. So what the heck is this workflow going to be doing? Well, like I said, this
00:01:36is research on steroids. So we are going to be inside of Claude code, and we are
00:01:40going to do some research via YouTube, right? My source of data in this case is
00:01:45going to be YouTube videos. To do that, we will use a specific skill. From there, we
00:01:50are going to send that YouTube data to notebook LM via Claude code. Notebook LM
00:01:55will do analysis on those videos for us. Notebook LM will also give us any
00:02:00deliverable we want, whether that's a podcast or a video or an infographic or a
00:02:04slide deck. And then it passes all of that back to us inside of Claude code. All of
00:02:09this is executed through skills. Furthermore, we are going to combine all
00:02:15those sub skills into essentially one super skill. We will do this using the
00:02:22skill creator, right? So that's where the skill creator comes in, and obviously the
00:02:26notebook LM stuff will come into play here. What about Obsidian, right? Because
00:02:31this is good in a vacuum, but like, we kind of want to supercharge this. I'm probably
00:02:35not just going to run this workflow one time. Well, enter Obsidian. All this data
00:02:40we analyze, and more so than the individual data, the way we attack the data, how we
00:02:46like our analysis done, what we want the deliverables to look like, how we think,
00:02:50all of that will be recorded by Claude code in a series of Markdown files, a
00:02:55series of text files that Obsidian will be able to take a look at because this is all
00:02:59going to happen in our vault. Now, looking at Obsidian right here, the vault's great,
00:03:03right? For a couple of reasons. For me as the human being, I have great insight into
00:03:06what's going on in my text files. I can click through the files. I can see how they
00:03:09link together and I get cool and neat little graphs. But more importantly, inside
00:03:13of Claude code, all those Markdown files are transparent to Claude code itself.
00:03:19It's easier when it's set up in this Obsidian sort of format for Claude code to
00:03:22find the things it needs. Furthermore, over time, we will be able to refine how
00:03:29Claude code speaks to us and thinks in this manner via the Claude dot MD file,
00:03:34which over time means Obsidian helps Claude code do this workflow in a manner we
00:03:41want, right? With Obsidian added into this workflow, we're able to turn Claude code
00:03:47into like this well-trained personal assistant that executes this workflow on our
00:03:53behalf. And that's super powerful. This almost becomes like a self-improving loop,
00:03:58right? Because the more I run the workflow, the more it gets its analysis in the way I
00:04:02like it. The more I talk to Claude code, the moral head data is recorded and Claude
00:04:07code continues to build and build and build over time this corpus of knowledge and
00:04:11evidence for how I like to work. And so that's how we get this like awesome symbiotic
00:04:16relationship and all these things kind of helping one another by combining Claude code
00:04:20with the skill creator, with notebook LN, with Obsidian, right? And you can see how
00:04:24flexible this is because this sort of workflow changes whether, you know, you know,
00:04:28we can take out YouTube could be PDFs, right? You can even take out the notebook LN
00:04:31piece. You could really have any workflow here, right? Insert whatever flow. But if
00:04:37you keep this template of flow Obsidian and improve skills via the skill creator, you
00:04:42have something super powerful at your fingertips. And it's not something a lot of
00:04:46people are doing. Now, before we get into how we set this up, exactly a word from our
00:04:50sponsor, yours truly. Again, if you want to learn more about Claude code, I just
00:04:56released a Claude code masterclass inside of chase AI plus. It takes you from zero to
00:05:01essentially AI dev regardless of your technical background or lack thereof. Chase AI plus
00:05:07is great if you're serious about AI and you're trying to make a career out of this
00:05:09thing. So definitely check that out. Also, there is a free chase AI community. You can
00:05:15find that in the description. All of the skills we talked about today, as well as a
00:05:18number of other free resources can be found there. So there's something for everybody.
00:05:23So first thing we got to do is create our skills. You will notice I am inside my vault.
00:05:27We have to be in whatever our vault folder is for Obsidian to pick up on this stuff. Now,
00:05:31skill creator skill, how to install it, get it working. Make sure you check the video
00:05:35above. I go in depth, but the five second version, you're just going to do slash plugin.
00:05:40You will search for the skill creator tool. You can see mine is installed right here.
00:05:46Skill creator, install it, exit Claude code, spin it back up. You're ready to go. And so
00:05:51if I want to build a skill, I'm going to do slash skill creator to make sure it actually
00:05:55uses the skill. And then we're just going to describe it. In this case, I said I wanted
00:05:59to create a skill that searches YouTube and return structured video results. It should
00:06:03use the YT dash DLP to search for videos by query, return the results, blah, blah, blah,
00:06:08blah, blah. This is how it is from a YouTube thing. Adjust it for what you want as your
00:06:11source. Again, these prompts will be available inside of my community. Once you run that,
00:06:15it will create the skill automatically inside of your dot Claude folder. It'll give you some
00:06:19descriptions about what it did with the skill creator tool. Remember, we have the ability
00:06:23to run tests on it as well if we want to, but we'll skip that for now. So that gives me the
00:06:28YouTube skill. I can now search YouTube. What about the notebook LM side? Well, just like
00:06:31the last few things, I have a full video, deep dive on that. Check it above, but I will give
00:06:35you the 32nd rundown. So notebook LM doesn't have a public facing API. So for us to connect
00:06:41Claude code to notebook LM, we are going to be using this GitHub repo, the notebook LM
00:06:46dash PI. I'll put a link in the description to install. It is very easy. We're just going
00:06:50to run these commands inside of our terminal. So we'll just copy this. I create a new
00:06:55terminal. Again, I am not inside of Claude code at this point. This is just purely the
00:06:59terminal and I will paste them in there and run the install. After I run that install,
00:07:03I need to log in to notebook LM authenticate. You see it here in the CLI section. So I just
00:07:09copy that notebook LM space, log in, put it in the terminal, hit enter. A browser window
00:07:14will pop up asking me to log in. I log in and that's it. You are done and installed and you
00:07:19can now use notebook LM. However, we need to teach Claude code how to actually use it. That's
00:07:24where the skill comes in. Now this repo gives us a command to do it. We can run this notebook
00:07:29LM skill install if we want. We also have an ability. What would probably be better now
00:07:34that we have the skill creator would be to like just copy, you know, essentially this
00:07:38entire GitHub repo or just put a link to it and give that to Claude code and say, Hey,
00:07:43use the skill creator to create a skill for notebook LM dash PI. And you see that prompt
00:07:50right here. Skill creator create a skill so we can best use the notebook LM skills seen here,
00:07:55right? Like this is like one of the best things about Claude code is it will do things that affect
00:08:00its own use, right? Like it understands how skills work within its own ecosystem. And so when I do
00:08:06stuff like this, it sort of self improves in a way, which is great. And once you run that, you'll get
00:08:11the same message essentially that you saw above when we created the YouTube search skill. And when
00:08:15it comes specifically to the note LM skill, these commands allow us to do anything and more from the
00:08:21Claude code terminal that you could do inside of notebook LM normally. So we have the ability
00:08:26to create our own notebook. We can add as many sources as we like. Well, up to 50, it could be
00:08:30from our drive, copy text files, YouTube, et cetera. And then like I mentioned before, we have all the
00:08:35deliverables that notebook LM can give us audio review, mind map, flashcards, infographic, et cetera,
00:08:41et cetera. So now we have the YouTube skill and this graphic has just become hideous, right?
00:08:45Let's clean this up. So we have the YouTube skill. We now have the notebook LM set up, but again,
00:08:50I don't want to tell Claude code one by one, or I do the YouTube skill, sick thumbs up. Okay. Now do
00:08:55the, do that skill. Cool. Thumbs up. I want to do this all at once. I just want to turn it into
00:09:00one skill and that's what we'll do now. We're turning our workflow into a skill. And so to
00:09:04create that YouTube pipeline, that workflow super skill, you can see same exact process,
00:09:09skill creator. And then I just did a stream of consciousness for it to create that pretty much
00:09:15saying, Hey, I want this YouTube pipeline skill. I want it to use a YouTube search. I want it to
00:09:21send it to notebook LM and I want, Hey, if I ask for it, some sort of deliverable and I want to
00:09:25brought back, right? That's what I said in way too many words. And at that point it will create the
00:09:30skill, tell you what it did, and then ask if you want to run any evals, which is up to you. And at
00:09:35that point, our workflow is essentially all set up, right? Skills are ready to go. It's inside obsidian.
00:09:41Now all we have to do is execute it. So let's do that. And in our use case, what we will ask for
00:09:47is we will ask for Claude code to go search up videos that have to do with Claude code and MCP.
00:09:53I want to find out the top five MCP servers. So I wanted to go grab the sources and I wanted to do
00:09:58analysis, not just what on the top five are, but how are those videos doing? Like what is driving
00:10:03views? What are some sort of outliers? What are the gaps and what can we do to capitalize on them?
00:10:09And I'll also ask for it to take that analysis and create an infographic for me. And that's the exact
00:10:14prompt you see here. I have my YouTube pipeline skill up and loaded. I could have used natural
00:10:18language, but anytime you use the slash command, you know, it's going to work a hundred percent.
00:10:22Like I said, YouTube MCP, Claude code analysis, and I asked for an infographic. So you can see it's
00:10:28starting the pipeline, calling the sub skills with notebook LM, as well as YT search. And again,
00:10:34the great thing about this notebook LM stuff is the fact that all of this processing by the AI is
00:10:41done by notebook LM. Like these are tokens you're not playing for and Claude code doesn't have to
00:10:45use. This is all offloaded to Google. Thanks Google. So after six minutes, the analysis is
00:10:50complete. Know that most of the time when you're talking about like, just like text analysis and
00:10:54you want to know what a notebook LM is giving back to you. That's pretty quick. The deliverables can
00:10:58take time. So if you're looking for a full slide deck, for example, that can sometimes take up to
00:11:0315 minutes, right? Cause it's several images it needs to create. If it's just like a one-off,
00:11:07like an infographic handful of minutes. So here's our infographic, right? Talking about MCP. Cool.
00:11:13We didn't give it a lot of guidance in terms of the visuals that we wanted to see, but solid, right?
00:11:18Suba base, context seven play, right? All right. Breaks it down into autonomous coding and the
00:11:23essential vibe coding stack. So what did they say? Suba base, Figma, Sentry, post hog, context seven,
00:11:30play, right? Can't argue with that. And then up top, you can see here, it gave us the full
00:11:36markdown file for the research. Now, remember this is inside Obsidian. So while this seems just like
00:11:41a normal markdown file where stuff is randomly in double brackets, it's a much more, it's much
00:11:46more obvious and easy for us as human beings to see this in context via Obsidian. Here's the same
00:11:51document inside of Obsidian key takeaways servers. It has the back links that will show me the other
00:11:57articles that's related to, I can see it inside of the graph, right? Cool stuff, but that's not
00:12:02where the Obsidian value ends. Remember the Obsidian value is the fact that I have, you can
00:12:07see it over here on the left, all these markdown files, which taken in the aggregate, pretty much
00:12:13show Claude code, how it is I work. And if we look over here to the Claude MD file, and that's what we
00:12:20see right here, the Claude MD file becomes that brain within a brain, right? If this vault is the
00:12:25second brain of mine where I have all these ideas, well, the Claude that MD file is again, the brain
00:12:30within the brain that tells Claude what this all means and what that means in terms of conventions
00:12:37of how to talk to me, how to give me deliverables, how I want things done. And so, like I said,
00:12:41over time, this vault will grow and grow and grow and grow, but it's very easy for Claude MD to grow
00:12:48along with it. And again, be trained and learn and grow alongside this corpus of knowledge. And it's
00:12:54as simple as telling Claude code, hey, update Claude MD based on our latest conversations.
00:13:00So these conventions are maintained and you're actually doing what I want to do. And that's as
00:13:04simple as saying, can we update Claude MD? So it better reflects my work style analysis and output
00:13:09preferences based on our latest conversations, right? Something as broad as that is enough for
00:13:15Claude to kind of like go nuts with it. If you want to be more specific, you can be more specific,
00:13:19right? That's the great thing about this is it's very flexible and it's up to you. And over time,
00:13:25that relationship between Claude code and Obsidian is what it's going to cause it to improve its
00:13:31performance, right? Doing that over the course of a week won't have too much of an effect. Doing it
00:13:35over a month definitely will. Doing it over a year and hundreds and hundreds of documents and
00:13:40conversations that will have a huge lasting effect. So that is where I'm going to leave you guys today.
00:13:46I hope you got more out of it than just this workflow in particular. And, you know,
00:13:50a little inside view of how I do my sort of content research, because again, the big sell here with
00:13:55this is that we can take all this away, right? And all we need is some sort of workflow in some manner
00:14:02that helps you, right? And whatever it is you do. And if we can take that workflow and turn it into
00:14:07skills and even turn a massive skills into a single skill and plug it into this sort of pipeline, well,
00:14:13then we get the situation where everything is helping each other, right? So, and again,
00:14:18on the long term, tons of value there. So let me know in the comments, what you thought as always,
00:14:25if you want to learn more about Claude code, you want to check out the Claude code masterclass,
00:14:28check out chase AI plus there's a link to that in the comments. And as always, I'll see you around.