NotebookLM MCP + Claude Code: Build ANYTHING!

Today we're going to be running through how to operate notebookm and connect it to claude. Like you can see right here. For example, in this example integration, we actually used notebookm and claude to use this NCP to generate for example audio. And you can generate whatever you want and automate pretty much anything that you can do inside notebookm and then connect it inside claw code which basically just empowers claw code to be able to create for example podcasts or for example infographics or even do deep research via notebookm and look through your previous notebooks inside notebookm using claude and it's fairly simple to set up. So I'm going to show you exactly how to set this up today and how to use it etc. So for example, I can access all of my notebooks right here using the claude MCP like you can see and then it can actually download the audio for us here with this link or we can just access it directly via here right and you can see how we can basically give this prompts inside the chat to generate audio or to generate videos and that sort of thing and there's a bunch of cool stuff you can do for example like deep research audio views adding URLs querying across any of our 122 notebooks over here. So, we're going to get straight into this and we'll start testing this out. So, how do we do this? What we're actually going to be using is a GitHub like you see right here, which is notebook LMCP server. So, this allows you to connect Notebook LM directly to Claude, right? Amazing GitHub by Jacob. So, what we're going to do here is we're going to go straight into Claude code. So, let's start running this. Open up Claude like you see. And then we can set up a folder or what you can do is you can go into your terminal just run claw code like and then from here we're going to continue and then we're going to plug this in. Okay. So we're going to accept the terms go ahead paste in the GitHub and say run this and that's going to begin to start setting this up. Now what you can actually see is the commands that you can use with these features. Right? So this is the MCP and the GitHub. So for example you can list all your notebooks, create notebooks, get notebooks, describe what you want. You can for example actually use this to generate audio and that sort of thing and you can use this inside for example claw code Gemini CLI and cursor or visual studio code as well and [snorts] it's fairly easy to set up. So what you can see is inside the terminal over here. This is beginning to go ahead. So we're just going to keep running this. I tend to find it's easier just to paste in the details from GitHub and then run it directly inside Claude Code. Whereas if you're a coder or something like that, if you're a developer, you can probably do this even faster and easier. So this is going to begin to install it like this. The other thing that we can do here is we can actually grab the MCP setup directly from my notes right here. So let's grab that. And if we go over to Claude like so, we're going to go to Julian Goldie settings. Then we're going to go inside developer and we're going to edit the config. Right? So if we edit the config which is the CL desktop config JSON we can paste in the details of our MCP server which includes notebook 11 over here. We hit save like that. Close claude and then reopen it. And you can see the notebook is set up. Now it has failed but we'll fix that in a second by using our terminal. You can see it's got to install a bunch of stuff first before we can actually start using the MCP. So that's what we're waiting for inside claw code. Now what it's going to do is ask us to sign in. So, I'm just going to do that here and then grab the password. Now, if we go back to Claude Code in the terminal, we're going to proceed with that. So, we've just logged in with Google and now we're waiting for claw code to install the rest of this. So, from here, we have to restart claw code and then we can go straight into this. So, if we go inside the settings, you can see that the notebook LMC is now running. Beautiful. We'll go to new session inside the chat here. And if we go to connectors, you can see notebook LM MCP is set up. So, let's test this out. And there's a couple other things we need to set up before we can start running this. But we're on our way. Boom. There we go. Right now, if we click on open notebook, you can see that is now connected to notebook. Great. So now what we can actually do is we could run deep research, we could generate content, we can query, we can add sources, we can create a new notebook, etc. Right? And this is all connected to our notebook, which is pretty amazing. And you can see here it even links directly to our notebooks so that we can just pull those up and share them and that sort of thing. Right? So let's test this out now. Let's go inside Claude here and we'll say create an audio overview. We'll have to allow this. So we're just going to click on and you can see it's now pulling in the notebooks and it's offered it's actually looked at the notebook itself and it said this is quite empty. Do you want to add more sources first or just try creating the audio for you first? Which one do you want to do? So we're going to say just create the audio just to test this out. So it's using audio create and now you can see this is beginning to generate the audio using notebook. Right? So we can basically control notebook using audio views like you can see. So if we have a look inside our notebook as you can see right here it's

generating the audio overview and that was fully operated by Claude. Now let's try something else now. I'm going to see what the limits are on this. So you can see here that's going to take about 5 to 15 minutes to generate. Let's try something else. So if we go inside claude code and we'll create a infographic. Now it says Claude wants to use infographic create from notebookmcp. Let's try that. And now you can see it's generating the infographic as well. Pretty amazing. Like you can actually use claw code and leverage the power of nano banana inside claw code. Now obviously claw code itself it cannot generate images but if you're using the MCP well now it can generate images. It can generate podcasts. can generate videos, right? And you can do it all inside the chat directly inside Claude locally as you can see, right? Pretty amazing. Now, if you go over to Claude on desktop as well, let's have a look here. It's actually given us the breakdown of what it's doing so far. So, for example, we got the infographic generating and we have the audio overview generating. As you can see, we can open up notebook as well. So, we can actually have a look what's going on. And then also if we click on the plus connectors over here inside settings you can see that we don't have the MCP directly inside the claw chat but we do have it inside the chat but not inside the settings. Right? So if you want to keep operating this just go back to claude directly on the desktop. That's the best way to use this. So let's see if that's generated. So that has actually generated the infographic but let's see if claw can pull it in now. So if we say okay just check back now and see if it can gather the status of the infographic. Boom. Both are done. Right. And what it's actually done is created this beautiful little table that's way easier than the UI of notebook. Right? So if you look at the UI of notebook, it's a little bit messy, a little bit cluttered. It's like there's these tiny little icons at the bottom. But if you actually go to Claude, the links are just much more neatly organized, right? And then you can see the infographic right there. Nice. All right, let's try something else. Let's see if we can create a video now. So I'm going to go inside here and say okay now create a video view with notebook lm MCP. See if we can do that. Boom. There we go. All right. Video view is creating as well. Right. And it's got the type explainer, language, English. You could also say create a video in Japanese using this notebook. I think as well you could do that. But the point here is it's quite easy to set this up. You've now got a full status of everything that you're doing. You can create audio, infographics, videos, etc. inside notebook LM using the MCP and you can operate it all remotely from claude. Pretty insane. And how long did that take to set up? Like just a couple of minutes. So if we go back to notebook here, let's just check if it's doing. Yeah, there we go. So we got the video view generating as well. And you could also pull in like any of your other notebooks. You could do deep research inside here as well. So if we go back to Claude here and we'll do deep research in the notebook for information about AI agents. Right, just pull in some more sources and now it can check the status and it's operating remotely for deep research inside notebook. Right, so you can basically control the whole notebook app using this method. Super easy and simple to set up. It's generate videos, infographics, audio. It can pull in deep research. It can analyze and communicate with your notebooks. But the cool thing here is obviously I don't know about you, but for me like claw code is 10 times more powerful when I'm using it and coding out with it. And so it's much better to use the MCP inside Claude than it is to actually use Notebookm directly. So let's see what we got on the questions here. So Jay says, "Bro, what's up? Use anti-gravity and claw and combination to come with smooth workflows 100%. Start devops says, wait, what? This is crazy. " So we can also connect the notebook LM MCP on Open Code or any other CLI. Yeah, I think you could also use it inside anti-gravity for example, too. John says, "Graphics look great. " And Sam says, "Mr. Consistent. Thanks for the help. Thanks very much for watching. What I've actually done is I've plugged in the full SAP and the links to the MCP server inside the AI success lab link in the comments in description. If you go to the classroom here, then you're going to go to January automations. Like you can see, you can get all the video notes and the step-by-step guide for installing this. What I'm also going to do is just paste in here the MCP settings. So, if you just want to paste this into your MCP settings just like I have, you can just grab this bit right here. There's a couple of other MCPs in there like for example Apple script which actually allows me to control Claude locally like all my documents and everything like that and also browser MCP which allows me to control my browser with Claude as well. You don't have to install those. If you just want to install notebook LM you can just install the MCP right here. Then follow the instructions I've given you today and you can get this set up yourself. Pretty simple and easy. So thanks so much for watching. If you want to get that link in the comments in description. This community is a free one that connects you with 44,900 people, loads of AI builders inside there. If you go to the classroom, then you go to January 2026 new automations and go to this section, you can find all the video notes from today, including how to set this up, the GitHub, and everything else. Right now, if you want to learn how to use Notebook LM inside your business to automate it to save

time, you can join the AI profit boardroom. This is an amazing community of serious AI builders and business owners who are automating their business with AI. And you can see for example, we have a daily accountability group. So you have the action takers daily check-in where you can post your goals for the day, what you're going to work on, what you going to build, etc. And then people actually check in and make sure you've done it right, which is amazing. You can also see that loads of people are winning with this stuff. So it's not just me. So for example, like Barb was saying, thank you all. She was struggling and she was blown away. So she said she was knocked flat by the warmth and caring of all the folks here in the AR profit boarding community. Right? So she built out some awesome stuff and said just wanted to thank Jud and Golden and everyone here. I'm grateful to each and every one of you which is amazing. We've got for example Jade building out his first AI video project. We have Maddie absolutely smashing it with AI and you got Steve Light who found a bunch of beta testers for his tool as well. So, if you want to get all of this, it's inside the AI profit boardroom. And if you check out the six week road map here, we'll show you how to go from complete beginner to expert in just 5 weeks with AI automation, plus how to build your first AI agent in under five minutes. On top of that, you'll learn how I automate avatar videos for YouTube. You've probably seen a few of those if you follow me on any social media platform. You'll learn how to get more clients for your agency. You can watch about the Q& A and coaching call recordings over here. You can get my best AI SEO automations and a sixe road map for building and growing a YouTube channel based on what's working for me. And inside the coaching calls section here, you can jump on four coaching calls per week. a live call with an AI automation expert, get help, get support whenever you need it.