I Found The Perfect AI Prompts For Working With Notes in Tana

What if, instead of just using AI to summarize your notes, you could turn it into a powerful thinking partner that challenges your ideas and helps you think in ways you never could on your own? Well, I developed an AI system that helps you do exactly that, and in this video, I'm going to share with you every prompt, command, and step-by-step setup that you need to create your own personalized thinking partner. By the end of this video, you'll have a completely new way of using AI to help you think better about your ideas.

Everything I'm going to show you today is part of my AI-Powered Note Taking Playbook, a collection of 50 of my best prompts for using AI to work with your notes, so I'll leave a link in the description for you to check that out. All right, so the prompts I'm going to show you today are all in my AI-Powered Note Taking Playbook. There are over 50 prompts in this Playbook around exploring your ideas, connecting your ideas, and distilling your ideas down.

They can be used in any LLM that you choose. Today, I'm going to use Tana and its inbuilt AI to show you how I use some of these prompts. So let's get into these prompts.

I'm going to show you five of my very favorite ones today—my go-tos. The first three are all about connecting your ideas, and the last two are about distilling your ideas. This is how I think about thinking, right?

So when it comes to an idea, what you want to do is you want to expand it. You want to pull it apart, get all of the information, and then distill that down to its very essence. That's how you get a really good idea.

These are the kinds of prompts that I use to do this. The first one is called the Five Question Framework, and it is my go-to, especially for getting started on a new idea. I find that, you know, like anyone, I don’t like looking at a blank page.

Even when it comes to kind of a spark idea that I have, I find a great way to get started with "What do I think about this? " is to have AI ask me some questions. That is exactly what the Five Question Framework does.

So you’re looking at my actual notes in Tana, and what I do is whenever I need to get started, I come up to my AI menu, and I've got all of my prompts in here ready to go. So let’s just quickly have a look at what the five-question prompt is, and then I’ll show you how it works. If I open this up, what you see here—the prompt itself—is this: “Ask me five questions about the topic or idea.

Your goal is to discover any opinions, experiences, or insights. Ask each of these questions one by one. Don’t let me sit on the surface; push me to go deeper.

Treat this like a conversation between two curious people. ” So that’s really the essence of the prompt, and we're trying to get AI to throw us those questions so we can start really thinking about things. If you’re interested in how to set this up in Tana itself, what it is is a command that you create, and the command module that you use is "Start AI Chat.

" This basically triggers the AI chat to open with the prompt already filled in there. Now, we have an agent in here which works kind of like a custom GPT, and if I open this up, we’ve just got a simple system prompt. If you think about a custom GPT, this acts like your system prompt in that custom GPT; it provides this information to the AI without you having to do it each time.

So I’ve just got something here: “You’re my thinking partner. Your job is to draw my personal, interesting insights. You’re curious; you like to ask questions.

Your job is to push me to think and encourage me to go deeper. Our outcome is to go from fuzzy idea to fully fleshed-out idea, and this is the idea that we’re working on today. ” Then, what you want to do to make sure that the AI gets all the context of your note is just add this little tag here: CIS for context.

Once that’s done, what we’re going to do is go to "Explore," and we’re going to open up the Five Question Framework. You can see we’ve got our thinking partner here, and it's starting to think. So, first things first, it starts with the first question.

The article mentions a second brain being like a rich, rotting compost heap of ideas. That was one of my favorite quotes from the article. “What’s your experience with building a second brain system, and how would you describe your own metaphor for it?

” Again, this is an easy question for me to answer, but it starts to get me thinking about what my unique idea is. All I have to do is just keep answering the questions. I've just answered the question, and you know, really not with very much info—just, you know, “I’ve been kind of doing this for about 3 or 4 years.

I used to have this really rigid…” you know, as I was thinking through this, I thought, “You know what? I used to really think rigidly about how to…” Take notes, and now I have a much more kind of flexible outlook on it. So then, you know, AI looks at that answer; it kind of knows what we're talking about, and it starts to then ask me, you know, secondary questions.

Um, which is like, "Okay, tell me more about, you know, what caused this shift? What made you realize the rigid system wasn't working? " And so then I just continue to keep, you know, answering questions and answering questions.

As I do that, things start to then come together for me. How that then works is I will take, you know, a whole lot of, um, you know, I'll take all of those answers and will start to then put them into my notes and start to organize them in that way. I find this five-question framework just a really great way to get my thinking started; it's really, really cool.

Okay, so that's the five-question framework. The next one is called the WTIF game. Okay, and this also helps me to think a little bit deeper and a little bit wider about things.

Okay, so let's play this game. So let's look at this one. Um, so if we have a look, if we open this up and we have a look at what the "what if" game is, all these prompts are set up exactly the same; they use exactly the same agent.

Okay, um, but what they do is they have a different initial prompt. Okay, and this particular one is, "Ask me as many what-if questions as you can to prompt my thinking. Explore different contexts, constraints, and perspectives to help me expand on the idea.

" Again, we're saying, "Ask me one what-if question at a time to give me time to think and answer without feeling overwhelmed. " Now, this particular note is actually just something that I was thinking about, and it's actually an audio note that I took while out on a walk about how to build effective habits when reviewing your notes. Okay, so what we're going to do is we're going to go to explore, and then the "what if" games.

Again, this opens up a chat, and the prompt's already given, and then it starts with a question. Now, you can tell I like questions when it comes to exploring different ideas. So, um, this one: "What if instead of having fixed review schedules, the system could intelligently suggest when to review specific notes based on their relevance to your current work or thinking patterns?

" Now, that is an interesting question because I wrote over here about different daily, weekly, and monthly kind of routines for your notes. But I love this idea of having something that is smart that actually knows when you should review something. So, this kind of brings up all kinds of sparky thoughts for me, um, and all I have to do is kind of start to write things out.

So I simply just, you know, input some more things around like, "Okay, this is what sparked for me when you said this. " And so then now we're going deeper, right? And so now, what if we think about notes having different urgency levels or review triggers?

Like, how would you envision a note knowing when it needs to be reviewed? And so it's really asking me to keep thinking deeper about this. This is something that I would not have thought of on my own, and this is why I love using AI for this process, because it does trigger you to think in different ways.

It's like having a person next to you who's saying, "Well, what if you thought about it in this way? " or, "What if it was this? " And it's really, really effective to start you thinking different thoughts.

So, the five-question framework and the "what if" game are great for getting you to really get into your ideas and start to think a little bit differently—start to generate your own unique ideas. The Devil's Advocate game, which is our third prompt, is designed to, uh, basically come up against you and come up against your idea and make you defend it. Uh, this is also one of my favorite games.

So I've got a note here which is called "Don't collect quotes; collect ideas. " Now, I've been thinking about this for a long time, um, and I actually have a short that I want to record, um, but I couldn't quite get it right, and I didn't know why. And so, um, I used the Devil's Advocate on this particular idea to really, really get clear on what my argument was.

Okay, um, so how this one works is let's just have a look at Devil's Advocate. So again, it's the Start AI chat module; uh, then, uh, we're using the thinking partner agent. So it's exactly the same as the other two, but the initial prompt is literally this: um, and this is why I love using this agent, because we can actually prompt the agent with, you know, really the bulk of what we're trying to do here, and then the initial prompts can be pretty simple.

Okay, so this one is literally, "I want you to play Devil's Advocate; challenge me on my idea. Don't hold back. " So all we have to do here is, uh, go to explore Devil's Advocate.

It's going to open up the chat, and then it says, "Okay, let's challenge this idea. " Okay, now this particular one doesn't ask it in single questions, um, but, you know, gives me some ideas to start playing with. Okay, so let's go to the first one: "Aren't you creating a false dichotomy here?

Why does it have to be either quotes or ideas? Couldn't quotes actually be the seeds that grow into ideas? " Now this.

. . Is a great question, and it's actually the question that I asked myself: "Why is it black and white?

" Right? Um, and so I can kind of continue to noodle on that. Second, you mentioned that storing information doesn't require cognitive effort, but isn't the act of choosing which quotes to save already a form of cognitive effort?

And I'm saying, "Hm, yeah, okay, maybe it is. " And so all of these questions are designed to get you to think outside your regular thinking, and that makes your ideas even better. And so "Devil's Advocate" is just a really, really fun prompt to get you to think bigger about your ideas.

All right, so those are the three prompts that help you to explore your ideas, to blow them out, to make them really big, and to collect as much as you can about those ideas. These next two prompts are distilling prompts, so they are about distilling your ideas down into the very essence of what they are. Okay, and it's a great way to figure out if you've really understood the idea.

So this first one is called "first principles. " Okay, so this is a note that I have on Zettelkasten versus commonplace books. I have been exploring this idea for a really long time, and I've got lots and lots of different notes here, okay, about all sorts of different things.

And so, how this particular prompt works is if we have a look at it: first principles is—we have our thinking partner agent, okay, and then we have our initial prompt. Okay, so this one is: "What are the first principles or fundamental truths behind this idea? What assumptions can I remove to get to the core?

" I also add here, "Remember, you're my thinking partner, so it's your job to help me get to the first principles, not give me the answers. " So again, you know, the beauty of these is that we're not asking AI to do our thinking for us; we're asking AI to be our thinking partner. Okay, so if I then come here and go to "first principles," then the chat's going to open, and we get our first prompt here, okay, and it's prompting you to keep thinking.

Okay, so let's start peeling back layers. What would you say is the most fun or human need or behavior that drives someone to keep a commonplace book? Now, that is such an interesting question, okay?

And that's different from just me having this prompt open saying, "What's the first principles of Zettelkasten versus commonplace? " Because it knows about the notes; because it's got all that context, it can really ask me very, very direct questions. You know, when you distill the methods, what's the core human activity happening here?

This lights me up straight away. Like, I want to stop the video and start writing about this straight away. So, this is just a really great way to, you know, get to the very core essence when you have a lot of information.

Okay, so when you've collected a really big idea, sometimes it can feel overwhelming—all of that stuff you've collected. You're like, "Where do I even start? Like, where do I even get to?

" And this strips out all that and basically gets you to the first principles really, really quickly. And this one is all about helping you do a 30-second pitch on what this idea is. It's a great way to be able to explain your idea really fast.

We can use exactly the same note, and you can see that you can use any one of these prompts on the same note and get different results. So what this one looks like is, if we go to the pitch, it's the AI chat module. We've got our thinking partner agent, and then the prompt is, "Help me summarize the idea in a 30-second pitch as if I were explaining it to someone with no prior knowledge.

You need to coach me to write the 30-second pitch rather than do it for me. " So, let's go. Okay, so it starts off pretty simply, right?

What do you think are the two to three most important points someone needs to understand about commonplace vs. Zettelkasten? What's the core message you want to convey?

Okay, now I could even say, you know what, actually, I don't know. So if I'm still kind of like, "I don't really know," like, "I need your help on it," then we just ask it, okay? And now we get a couple of really good questions, right?

So, what's the main problem or challenge that commonplace books solve for people? When you think about someone who will benefit from a commonplace book, who are those people? You know, what's the simplest way you could explain what a commonplace book is?

Right? And so then you start to kind of get into, you know, how do we really distill this down? Who is this really for?

You know, those kinds of questions. And this is a really great way to build a very concise way to describe your idea. I usually use this prompt when I am coming to the end of like an idea that's ready to go, and I need a really simple way to kind of convey that to the world.

So, those are my top five favorite prompts for using AI to work with my notes as a thinking partner. They are really, really powerful as a trigger to get you started thinking. Okay, so a lot of the time, before I used AI with my notes, I felt like I was stalled.

Okay? I, like, before I even got started thinking, I felt like this is too hard; it's too much. Of a heavy lift and by having these really simple prompts, and just being able to click that button and get a question or, you know, be guided through a process, it really does help.

I've just found that I love to explore my notes even more because it doesn't feel like that heavy lift anymore. It feels like I have a partner sitting right next to me, kind of guiding me through creating my own ideas and creating my own knowledge. If you like these five prompts, I have 50 more of them in my AI Power Note Taking Playbook; I highly recommend you check it out.

Again, as I said, you don't need T to make it happen. The prompts can be used in any LLM. You can set them up as a custom project in Claude, or you can set them up as a custom GPT in ChatGPT.

If you do use Tana, then I have a command that you can install that has them all installed and ready to go. Again, if you want to grab these prompts and more, then check out my AI Power Note Taking Playbook in the description. You'll get access to all the prompts that you've just seen, plus 50 more, so you can use them in any LLM tool.

If you use Tana, you can install the custom command menu directly into your workspace to get started. Now, if you want to explore more about my note-taking methods, then you might like this complete walkthrough of my note-taking workflow. And as always, like this video and subscribe to the channel if you want more great content like this coming up in your feed.