NotebookLM amb Obsidian i agents: què és oficial i què no

Notebook LM just got a powerful free upgrade and this changes everything about you how you use it. Right now most people are using Notebook LM the wrong way. They generate something, it sits there, they never use it again. That ends today my friend. I'm going to show you how to plug Notebook LM into an agent operating system that organizes everything automatically, turns your research into ready to post content with one click, and gets better every single time you use it. We're talking about videos, podcasts, infographics, slide decks, all in one place, all generated for free. And there's one thing inside the setup that most people completely miss, which I'll show you later, that makes this 10 times more powerful than anything you've ever tried before. Stick with me till the end and I promise you Notebook LM will never be the same again for you. Let's get into it. Today I'm going to show you a powerful new upgrade that just came to Notebook LM and this is a new way to set up Notebook LM so that you can get the most out of it. So, the powerful thing about this is, for example, if you're using Notebook LM the old way, it's very difficult to see everything you've created, it's very difficult to manage all of your different notebooks. And then also, if you create anything, like how do you make it better, right? So, for example, you can generate reports, you can generate videos, you can generate audio views, etc. But how do you make that content better, right? This is what you can actually do with HyperFrames. And this is new way to take the content that you have from Notebook LM and then make it 10 times better. So, you can see an example right here. This is actually a video we created with this new setup. And basically, what this allows you to do is edit anything that you create with Notebook LM. And so, it's a really

powerful way to start using this stuff. And also, it's free, right? You can use the Hermes video agent setup for free with Notebook LM. So, Notebook LM is a free tool, Hermes is a free open source tool, and the HyperFrames setup that I'm talking about today is a free open source project as well. And so, this bridges everything together to just make it more powerful and more useful, right? And so, you can also, this is a great thing about it is like you can have all of your Notebook LM notebooks, the audio views and the podcasts you create, all the chats, all in one place, all synced to your memory system, and then you can create whatever you want, and you can organize that as well. And I'm going to talk to you about why I did this as well. So, previously like turning research into useful content was a nightmare, right? So, for example, you'd have to go into Notebook LM, you'd use different sources, hit generate, and then you get something you like, but then it just sits there, right? To turn it into something else more useful, that's quite a resource-intensive and also very difficult to organize. So, what I've actually done is I've built an agent authoring system around Notebook LM, right? And this way, when I'm building this out and we're using this whole system as you can see, number one, we've got everything organized in one place. Number two, if we need to generate anything else, for example, you can see this notebook here that we've pulled in, we can organize that. So, we can generate whatever we want, we can generate a new AI video overview, or we could create a new podcast here, we could generate an infographic, etc., and it will pull it into this section. And then also, the cool thing about this as well is we can plug that into the video agent section here, right? And so, the

great thing about this is when we're setting this up, we've got it all organized in one place, and it's much easier to just use the stuff that we have and make it way more useful with Notebook LM, right? That's the difference. And you can see some examples of like people using similar agent authoring systems here. There's so much you can do with this, right? And so, I've shown you the proof. If you're thinking about this, setting this up, definitely do because it changes your whole workflow setup, right? And it's easy to set up, and also it saves a lot of time as well. So, the framework that I've created for this is something called the Goldy Knowledge Studio. And by the way, if you're watching this and you have any questions as we go along, feel free to ask. But basically, this allows you to take like raw knowledge into powerful, ready-to-post content, right? And so, this is like a pipeline where you use your AI agents to take the content from NotebookLM and make it 10 times better and make it 10 times more useful, right? And so, how does this work? There's five layers. Number one is the capture section here, right? And so, you've got a library of notebooks like you can see over here, and these are all plugged into your agent operating system. If you don't know how to connect your AI agents to NotebookLM, you can use an MCP, right? So, for example, this is a really good open-source MCP that I've used previously to set this up. If you don't know how to build like a nice UI like this, you literally just ask your agent to build it for you, right? So, you can ask Claude or whatever you want to build this out for you. From there, you can generate the content, right? So, based on that notebook you've created here, so for example, like this one, we can then go to the studio and we

can generate whatever you want, right? So, if we want to generate a podcast, we can select audio view, hit generate, and then it creates a podcast for us, like you can see right here. So, it is a much nicer, neater, organized workflow, and also you can customize it however you want. And you can generate all these different types, right? So, you can see this one is working right here, but we could generate a video or we could generate a presentation, a mind map, an infographic, a flashcard, a quiz, table, report, whatever you want using this process right here. Pretty amazing. And then from there, you can save everything automatically to your memory, right? So, everything that we generate with this system actually syncs to our Obsidian vault, right? Obsidian is a free memory system that you can use for AI agents. So, you see this knowledge graph right here where everything is linked together, and we've got all these notes. That's because my AI agents are taking notes on me and then exporting that to my Obsidian vault, right? Now, Obsidian is this tool that you can see right here. You can use it for free, and also it keeps everything organized inside a beautiful knowledge graph here. And also, my AI agents will write notes into this system daily, right? So you can see for example, we have all these notes on my community right here. If you keep scrolling down here, we have for example notes on how we use Claude and everything else. And also it shows you where it connects to and it tags everything and organizes it neatly too, right? So everything is just organized in one beautiful place. If we go to for example my agency called Agency, you can see all of the different systems and the way this is tagged into our knowledge graph right here, right? So anything that we create with Notebook LM also saves to the Obsidian vault, too. And

then the great thing about that is that when we create content in the future, it's all inside of all, right? So if we go back to Notebook LM over here, we check our assets, everything is saved in one place, beautiful. And then if we create anything, for example like we just created that audio of you, right? So if we go back to this, check out the studio, we've got that here. And once that's done, we can pull that into our vault like you can see here, right? And that makes it much easier to organize everything and keep it in one place. And then when you're ready, you can post this content, right? So for example if I generate a infographic or something like that, I've got it ready to go and I can just post that to social media if I need to, right? Or if I want to. And that's a full five-step process. Now, here's the thing that most people don't know about and this part really matters. So Notebook LM's always been good at generating content, but this new upgrade where you have it sit next to Hermes video agent inside Agent OS is what changes the workflow, right? So before, if you were just using Notebook LM directly, you might have your audio of you or a video of you, but you never really used it, right? It was just kind of stuck in one place. Now what you can do is you can actually edit it with hyperframes. So here's an example, right? We've got different examples right here of videos we've created with hyperframes. We've got this video of you and we can edit that with our AI agent to make it better, right? Way more powerful, way better setup. And so Notebook LM can create like the idea and then the video agent turns it into something amazing. Now, some people say, "I can already use Notebook LM directly. Why would I need

something like this bridge that we set up, right?" This is what turns it into something way more usable. So, using Notebook LM alone, you generate an artifact, for example, and it just sits there, right? You'd have to manually you'd have to manually remember where it is. Everything doesn't really sit in one place. It's very hard to organize. If you switch to ChatGPT or Claude, you're not really going to have everything linked Whereas, for example, with the agent operating system, everything is saved in one place, much easier to organize, much easier to just switch between my AI agents and have everything living next to each other, right? So, open Claude, Claude, Hermes, everything is right there ready to go together. That's the cool thing about this. So, let's talk about the old way versus the new way. This is two completely different ways of doing this, right? The old way was like you would use Notebook LM alone, right? So, you'd open up Notebook LM, you might generate an audio preview, you play it once, you forget it, right? And then, you forget about it, you don't know where all your files live, you forget about stuff. And the problem is like with that, you get isolated artifacts, you get everything scattered, and you don't really create anything useful, right? That you're actually going to use day-to-day. Whereas, for example, the new way with the knowledge system is number one, it's free cuz Hermes agent, hyperframes, and Notebook LM are all free. You can drop the source into Notebook LM, you can hit generate, everything syncs inside the vault as you can see over here. If you need to create more, you can go to the studio based on the notebook that you want to create for, as you can see here. And you can also see the status of everything. So, for example, this one is in progress now. And then, one click can create anything you want, right? That's a

powerful thing as well. And also, what we've set up here with the video agent is we actually have a AI avatar section here, right? So, we can create even more useful stuff by implementing my brands inside this as well. All right. And the result is a pipeline of that flows into one place where you've got everything in one screen, all organized for free, and it's just way more powerful and useful. Now, you could use this for audio views, for videos, for slide decks, mind maps, infographics, flashcards, quizzes, data tables, and reports. And some people say this might feel generic, right? The content that you create. But actually, if you have a way of customizing that using the video agent, then it feels a lot more like you, right? If I plug in this avatar into the system, how much better is that going to look? It's going to look 10 times better, right? Because it's unique to me, it's specific, it's more relevant to me and my brand, right? That's the difference here. And then other people say, "I'll keep things in NotebookLM. Why would I use Obsidian?" And the thing is, if you use NotebookLM, it doesn't really talk to your other tools. But Obsidian links to all of your AI agents. So, the whole reason for this is that Hermes agent can organize your Obsidian vault. And then, when you plug your Obsidian vault into any other AI agent, everything links together, right? The agents can understand your memory, your system, how it works all together, which is exactly what we want, right? So, that's basically it. That's how to create everything in one place. The one thing that I would say is the most powerful here is Obsidian as a memory vault, like super powerful stuff. And then also, with Hermes video agent, that's really cool as well. Now, you can just create much better stuff. And when you link it to NotebookLM, and you can generate free videos, or you can

generate, for example, like this podcast here. This is way more powerful, right? Way more useful. So, everything links together in a beautiful system. Josiah says, "Hi Julian, good to see you live. Thanks very much for joining." If anyone has any questions, by the way, feel free to ask as we go along. But yeah, that is the full system. Now, you get the If you want to set this up yourself, you can. If you want to get my full setup with the agent operating system and everything else, you can get that inside the AI Profit Blueprint, right? So, you get the full knowledge studio and the agent OS setup. So, you get NotebookLM already be into this agent operating system. You get the Hermes video setup as well. You get the workspace memory with Obsidian. You get the 30-day playbook and everything else inside the AI profit border, right? Now, the other great thing about this is so much more useful because Notebook LM alone is useful. 100 video agent itself is useful. But when you put them in the same dashboard with shared memory and one-click setup, the workflow changes completely, right? It becomes way more powerful, right? If I just want to generate a video, this how easy is that? I just go inside here, click generate, boom. That's how easy it is. And so, there's many benefits to this. Number one, you have a shared vault across every agent, right? Because Notebook LM is saving locally to your Obsidian file. The same with Hermes, Open Clore and Clore. And so, you stop having to manually organize everything. Number two, you've got one dashboard with one tab, right? So, notebook sits next to video, sits next to studio, studio sits next to goals. The whole pipeline runs inside one dashboard, right? SEO goals, SEO video, studio, notebook, even a Kanban board here for multiple teams of agents to put this stuff together. Additionally, as you use this system more, the better it gets, right? Cuz

it's self-improving. If every time you use this, Hermes is in a self-improving loop, which it is. It's automatically set up to do that. And then also, you're improving the system, making it better every day based on what And then additionally, you're improving your memory system. Guess what? This stuff gets better and better every time you use it. Whereas if you go inside Notebook LM, but you don't have a system and you're just using the defaults, it doesn't get better every time. You have to kind of wait for Google to make it better for you. And even then, it doesn't have context on you and who you are and what you do and what you did yesterday and everything else. And again, this is all free, right? You can ask Hermes to build the agent operating system for free. There's a dashboard for you. Now, you just give it the directions. Hey Hermes, create this dashboard for me. Here's how I want it to look, blah blah blah. Host it locally, blah blah blah, right? And then, Notebook LM itself is free. The MCP to link them together is free. And Obsidian is free as well, right? So, that's basically that's the whole setup. Now, some people say Notebook LM is too basic to really create a good content system. It's basic on the surface, but if you wire it into a powerful agent operating system, it's the input layer for the entire system, right? And so, Notebook LM is the engine, but the dashboard is what turns into the vehicle, right? And it's a completely different vehicle once it's plugged into the agent operating system. Other people say, "I'll never use like nine different types, right?" But, the thing is, you might have some podcasts you need to create, some infographics, maybe you got a slide deck to create one day, and so, just having that optionality is super powerful. And then, also some people say AI content isn't very good. But,

actually, if you have the video agent plugged in, if you have your avatar plugged into the system, if you have, for example, the memory system that's unique and personalized to you, all of a sudden everything becomes less generic and more useful. And then, we've got a question here, which is, "Do you know how to create an AI agent for free?" So, there's two options here. So, number one, you can actually just, for example, use an open source project that's ready to go, like Hermes, as you can see, right? That's one option. And then, also there's a really cool system here that I used. I've actually got a tutorial about it. It's called Build Your Own Open Claw, right? And so, you can create your own AI agent like this as well. Again, it's open source. I think it's built using Pi, which is what Open Claw was originally built on. And then, you can customize your AI agents exactly how you want them. And then, we've got a question from Don, which is, "Can you walk through how you have anti-gravity plugged in? Right, how is that set up?" So, anti-gravity is actually plugged in to the agent operating system. We are going off on a crazy tangent. It's plugged into the agent operating system using CLI, right? So, if you have the CLI, it's basically like a terminal that you can talk to your AI agents with. Google just released anti-gravity CLI, and that's how you can get this plugged into the system, right? So, if we wanted Hermes to build out an agent operating dashboard like this, we'd ask it inside the chat. If we wanted to add anti-gravity, we'd give it the CLI details and ask it to add ask it to add anti-gravity as a CLI inside there. Can you do a tutorial on how to use obsidian? Yeah, so I actually have that over here. If you go to the classroom inside the AI Profit Bot room and then

you go to the infinite context engine. This is the full system that we've used, right? So if we got a full video tutorial walking you through. And basically it's a combination of three tools, right? So we have obsidian and then we have So we we've got obsidian and then we have Omi and then we have that link to our AI agents and that's how it all works together. Matt says hi, good to you. And then we've got another question from Don which is I hear Google is about to stop the CLI. So basically what they're doing is they're switching anti-Google Gemini CLI to anti-gravity CLI, right? And then anti-gravity will be their main coding agent. So it's not like you're losing access to the CLI, it's just switching from Gemini which it was previously and then on June the 18th it'll be fully phased out and it'll switch fully to anti-gravity. So that's why I've set up anti-gravity now because it's less than a month away and we want to get that set up right away, right? So that's basically it and that's the whole system. Just to recap on what you've learned today, right? So you learned how to connect Notebook LM and Hermes for free. You learned how to set up a whole system that you don't need to keep re-explaining to. You learned how to organize all your files and all your outputs into one nice place, one nice dashboard. You learned how to create the video agent too using hyperframes so you get Hermes agent set up hyperframes for you. You learned how to organize it locally and then you learned how to link it to your sit memory system as well, right? So if you want to get the full knowledge system for me, Notebook LM is the engine, the Hermes video agent is the vehicle agent operating system. It's what turns it into a full studio that you can live inside and all your agents

can live inside as well, right? Now inside the AI Profit Bot room you get the full setup, the install, the coaching course, the community and also access to me as well so you can DM me and ask me for advice anytime. The So, inside there you get the full Agent Operating System installed. So, you get Notebook LM and everything else set up, right? If you actually go to the Agent Operating System here, inside the app of volume, you can see we have a full video tutorial, the last updated date, and a full zip file on how to set this up with your AI agents. You also get the Notebook LM system, the Homies video agent we built inside it, a 30-day playbook on exactly how to implement this, weekly live coaching calls where you can jump on these calls, ask questions to me, the community, share your screen, etc. And we have 3,100 members inside here, right? So, there's loads of cool people you can connect with. You can ask questions inside the community. I answer questions inside there every day. And then inside the classroom, you can also check out all of my new trainings and learnings. On the calendar, you can actually jump on those coaching calls, as I mentioned a second ago. And also inside the map, you can actually connect with people in your local city who are building AI agents and AI automations just like you are, right? So, you can see like all over here, we have people in pretty much all the main cities that you can connect with locally who are building AI agents like you. So, thanks so much for watching. Hope to see you on the next one, and cheers. Bye-bye.