Dan Martell - Full Claude Guide： Beginner to Pro in Under 15 Minutes

00:00:00.000 I've spent over 1,000 hours inside Claude, and I use it every single day to build tools,

00:00:05.800 run workflows, and even launch million-dollar companies. And look, you might feel super

00:00:09.820 productive using Claude right now, but what if I told you you're barely scratching the surface?

00:00:14.300 In this video, I'm going to walk you through every level of Claude user, from the amateur,

00:00:18.820 all the way to the one who builds fully autonomous systems that run without lifting a finger.

00:00:23.560 Let's start with level one, the amateur. See, the amateur treats Claude like a fancy Google search,

00:00:29.840 one question in and a response back, and then they close the tab. There's no memory, there's

00:00:34.640 no projects, Claude has no idea what you're working on, and you're using maybe 5% of what

00:00:39.660 Claude can do. It's like having access to a NASA supercomputer and calculating 2 plus 2.

00:00:45.940 So if you're in this level, here are two pro tips to help you level up. Number one,

00:00:49.840 make Claude interview you first. For example, you can tell Claude, before you answer, ask me any

00:00:55.100 questions that you need to perform this task properly. And watch it ask you the questions

00:00:59.840 context that it needs to give you the best answer. The second thing is you've got to make Claude check

00:01:04.340 its own work. Watch it catch its own mistakes, which is annoying because you think it would do

00:01:08.180 it in the first place, but it will give you a better output. You can use this every day by just

00:01:12.380 saying check your work. Now, before we go to the next level, we have to stop working from scratch

00:01:17.920 every time. So how do you make Claude have persistent memory? Level two, the regular. The

00:01:23.580 regular treats Claude like a workspace. They don't just chat with it. They use projects, which is

00:01:28.500 feature within Claude for their role. They use it per initiative. They use it per client. They're

00:01:33.880 using it per workflow. And what's cool is Claude finally remembers who they are every time they

00:01:38.420 come back. So the work gets better. And this is how you set them up the right way. So number one,

00:01:43.580 on the left side, you will see create a new project, click that, and then name the project,

00:01:48.340 the role you work in. And in this example, just say marketing. Okay. That's your role. Second,

00:01:52.500 we need to build a master prompt for your role. So every time you chat with it, it knows who you

00:01:58.240 are and what you're trying to accomplish. Now, the cool part is you can ask Claude, type this in,

00:02:03.180 interview me to build a master prompt for my role as a marketer. Watch Claude ask you questions,

00:02:08.960 you answer it, and then you get a file. That is your master prompt. And what a master prompt is,

00:02:14.100 is a file of instructions that tells the AI everything about you, how you like to work,

00:02:20.260 what your team looks like, what tools you use, everything about your role, so that every time

00:02:25.180 it gives you an answer, it can use that information to guide its output. The third is we've got to

00:02:30.100 add the files. So now that we have the master prompt, let's go back into the project, add that

00:02:34.060 as a project file. Then any other documents you have within your company, maybe it's sample data,

00:02:39.600 maybe it's examples, maybe it's files, maybe it's processes, put that in the files directory with

00:02:45.020 your master prompt. Now you have a customized space that has memory, that has context, and has

00:02:50.560 your specific workflow for how you like to work. So for example, every time I sit down to create a

00:02:55.800 YouTube video, I have a project folder to help me ideate and give me outlines and strategies. In my

00:03:02.060 files, I gave it everything that it would need. I gave it my voice document, my branding document,

00:03:07.700 my examples of previous scripts that I like. It has everything that it needs to create outlines

00:03:12.820 based on how I've done them in the past. And then I just direct it and then I hit enter and Claude

00:03:17.180 gives me the outlines. Here's a pro tip. If you want to take projects to another level, do this.

00:03:22.420 Ask the AI to interview you to create a system prompt for a specific workflow that you're trying

00:03:27.560 to create with your chats. The master prompt is like your ingredients. That is everything about

00:03:31.700 you, your context, et cetera. The system prompt is the recipe. It's the instructions. It's the

00:03:36.540 process. So you probably already have a process. You just never documented it. So the AI can

00:03:40.980 interview you. So you get that as the output. So once you get that, just copy and paste the system

00:03:45.620 prompt into your custom instructions and watch this thing run like a machine. That way you have

00:03:50.660 the same quality, the same format, every frigging time you use that project. And look, if you want

00:03:56.100 to go even deeper and how to incorporate Claude or any AI into every department within your business,

00:04:01.080 I have my AI company OS, the exact playbook I use to integrate AI into all my businesses,

00:04:06.220 and I'm giving away for free. If you want to just DM me YouTube OS on Instagram and I'll send it

00:04:11.540 over. So now your projects turn Claude into a workspace, like a desk with a process and the

00:04:16.940 whole thing's automated. But what about everything else? Think emails, calendars, work you have to do

00:04:21.860 in the browser. Level three, the integrator. The integrator plugs Claude into the tools where work

00:04:28.680 actually lives. We use Gmail, Calendar, Drive, Slack, Notion. The integrator connects it to Claude

00:04:36.640 so they stop bouncing around from all these different tools and tabs. Why would I have to

00:04:41.020 go copy and paste an email into Claude when I can just tell Claude to go get the email. The cool part

00:04:45.980 now is that you can build an act inside Claude. The integrator doesn't leave the tool. They do

00:04:51.920 the work in the tool and then they send the information that the tool created to wherever

00:04:56.320 it has to go. So here's how you operate at the integrator level. Number one, we use connectors.

00:05:01.220 Okay, so go through and connect your Gmail, your Drive, your Slack, your Notion, all the systems

00:05:06.280 that you have. It has an ability for you to connect those data systems so that it'll search

00:05:10.480 those places next time you're chatting with it. Two, build visualizations. This is my favorite

00:05:15.120 feature in Claude is that I can get it to visualize things. I can create graphs, bars, mockups, all

00:05:20.620 inside the chat. And even better, the third is build interactive artifacts. These are like little

00:05:26.220 mini apps that you can have outputs that are clickable and sliders and buttons. And it makes

00:05:32.060 learning or visualizing information or interacting with data so much easier. And here's an advanced

00:05:37.360 move that nobody talks about you can actually paste text into chat and tell it to put it inside

00:05:42.500 composer which is like a google doc inside of chat and then you can edit it and play with it so that

00:05:48.600 it gets the output exactly like you want my clot is connected to my google drive to my slack to my

00:05:55.240 calendar to my email so i tell it all the time hey look at my calendar tell me what i'm missing look

00:06:01.180 at my email tell me what i need to know i'm the ceo of my company scan slack for the last week

00:06:06.440 and give me CEO-level understanding

00:06:09.160 of what's going on across the company.

00:06:10.980 These are things that I run almost on a daily basis,

00:06:14.140 so I'm always on top of it.

00:06:15.880 Now, if you wanna level up with a pro tip,

00:06:17.900 install Claude in Chrome.

00:06:19.840 I use this all day long.

00:06:21.760 Oftentimes, I'll have a chat asking me to do something.

00:06:24.160 I say, write the instructions for me.

00:06:26.060 I copy, I paste it into Claude inside of Chrome

00:06:29.280 within some site, and I hit enter,

00:06:31.480 and it goes and does my work for me.

00:06:33.100 Now, this is all super cool,

00:06:35.020 But the truth is, Claude's just helping you at this level.

00:06:38.440 The next level, that's where Claude actually does the work for us.

00:06:41.940 Level four, the operator.

00:06:44.100 The operator, this is where you stop being the doer

00:06:48.460 and you start becoming the director.

00:06:51.100 I consider this the human in the loop,

00:06:53.640 where the operator, you, are now setting up tasks that run on their own.

00:06:57.920 You just review things and you approve it.

00:06:59.940 This is where you stop using Claude

00:07:01.760 and you start deploying it to solve problems for you.

00:07:04.520 And these are three ways any operator can deploy Claude

00:07:07.920 to do work for them.

00:07:09.020 So number one is system prompts.

00:07:10.780 Essentially, you want to have Claude interview you

00:07:13.900 to create these system prompts for any type of output.

00:07:17.260 So I believe that the future of intellectual property

00:07:20.340 and what makes teams or companies valuable

00:07:22.840 is that they have system prompts defined

00:07:24.940 for the things they create.

00:07:26.880 Level two is skills.

00:07:28.360 You will notice over time,

00:07:29.760 as you keep doing these things, these workflows,

00:07:32.300 you can actually save them as skills.

00:07:34.400 Cloud has a bunch of skills that you can install.

00:07:36.540 They've created hundreds,

00:07:37.820 everything from financial skills and marketing skills,

00:07:40.060 et cetera, but you might have some proprietary things

00:07:42.540 you do inside your company.

00:07:43.960 If so, you wanna create a slash, name it and run it

00:07:47.180 and then make it a skill.

00:07:48.560 My rule is if I'm doing it more than three times a week,

00:07:50.960 I might as well have as a skill.

00:07:52.140 For example, I'm always wondering

00:07:53.640 what's going on within my company.

00:07:55.060 So I created a company status skill

00:07:56.660 that analyzes all the analytics and the metrics

00:07:58.980 and the reports and everybody's updates

00:08:00.560 and then gives me a cute, concise little update

00:08:03.460 on the company as it is today.

00:08:05.460 So I just type forward slash company status

00:08:07.400 and it does it every time.

00:08:08.700 Third is scheduling cowork tasks.

00:08:10.780 Now, if you haven't heard of cowork,

00:08:11.960 that's an app that runs on your computer

00:08:13.580 and it can literally take over your computer.

00:08:16.140 So you can get it to run a job,

00:08:18.060 like migrate from this system to another system,

00:08:20.960 hit enter, go have dinner, come back and watch it get done.

00:08:25.400 So what a great operator does

00:08:26.800 is he looks at his system prompts

00:08:28.300 and he looks at his skills and he goes,

00:08:29.820 hey, are these repeatable outputs

00:08:31.940 that I can schedule and co-work to have it done every day.

00:08:34.800 So for example, every night I get a message

00:08:37.560 at eight o'clock that shows me my following day

00:08:39.920 and it looks at my emails and my calendar

00:08:41.920 and lets me know everything I need to do

00:08:43.460 as if I had a chief of staff and I do and she's awesome,

00:08:46.000 but I set it up because I don't need her to do it.

00:08:47.620 Then I know what tomorrow is gonna hold.

00:08:49.120 So if I have to make any last minute changes, I can do it.

00:08:51.320 I set it up once and it runs it every night

00:08:53.560 and I never have to think about it.

00:08:54.960 The operator realizes with AI, they direct the work.

00:08:58.840 They don't just do it.

00:09:00.020 now this is a massive pro tip and it's called chaining your skills so for example you might

00:09:05.700 have a copywriting skill that writes in your voice and then you're writing emails which is

00:09:10.260 also its own skill and it'll use the copywriting skill and those two skills might be called in a

00:09:14.980 completely different skill that's all about automating your inbox and those skills are

00:09:18.820 like separately packaged little genius agents that all chain together to create an outcome the whole

00:09:24.020 pipeline is chained skills. The operator is now really using AI. So schedules and skills,

00:09:30.660 those automate with your existing tools and connectors. But what if the tool doesn't exist

00:09:34.780 yet? Level five, the builder. Not too long ago, we shut down the company for two days. We held

00:09:41.180 an AI hackathon. We taught everybody how to code. In those two days, everybody on the team became

00:09:48.160 builders. Essentially, what you can do is use Claude to write code, build custom apps, dashboards,

00:09:55.060 internal tools, all inside of Claude Code. Now, Claude isn't just answering questions anymore

00:10:00.060 or setting up skills. It's literally shipping software. In this level, we're introducing a

00:10:05.540 new level of Claude called Claude Code. And just so you know, if you're in this level already,

00:10:10.220 you're in the 0.04% of the population. Most people don't know this. Inside Claude Code,

00:10:16.980 you can use three categories of building.

00:10:19.920 The first one I called loops.

00:10:21.260 These are reoccurring jobs that do work.

00:10:23.660 They're similar to co-work,

00:10:24.900 but they run inside of a server

00:10:27.340 and they can talk to other agents.

00:10:29.260 They can talk to other systems.

00:10:30.500 They can use other APIs and they're way more advanced,

00:10:33.360 but they're loops because they're always running.

00:10:35.700 Number two is tools.

00:10:36.620 These are things you build for one-off situations.

00:10:39.660 So you might build a tool for a project.

00:10:41.560 You might build a tool to help you accomplish something.

00:10:44.060 It's not gonna be something you're gonna do forever.

00:10:45.520 and that's why I consider them disposable.

00:10:47.540 The third level is apps.

00:10:49.200 I'm talking real software.

00:10:50.760 Everybody that was involved in that hackathon I talked about

00:10:53.160 built real working production software.

00:10:56.500 For example, I have this incredible person named Betty. 0.70

00:10:59.200 She's my house manager. 1.00

00:11:00.260 She manages our lives for my wife and I. 1.00

00:11:02.540 She built a system that manages all of our workflows, 0.92

00:11:05.780 every aspect of my personal life,

00:11:07.320 the cars, the real estate, the investments,

00:11:09.680 the budgets, everything.

00:11:11.120 And she's not a programmer.

00:11:12.580 in the world of AI, English is the new programming language.

00:11:17.120 Now, if you want a pro tip,

00:11:18.540 100% of the time before I build anything,

00:11:20.520 I use a feature called plan mode.

00:11:22.360 Essentially, you hit forward slash plan, enter,

00:11:24.560 and then you just dump your idea.

00:11:26.660 It hears everything you said,

00:11:28.000 and then it will ask you questions if it needs,

00:11:30.080 and then write a whole completed plan

00:11:32.080 that you approve first before it writes any code.

00:11:34.640 The reason why plan mode is so important

00:11:36.560 is because people always complain,

00:11:38.040 oh my gosh, Dan, it costs so much

00:11:39.560 to do all these AI apps now.

00:11:41.140 it's because they didn't play it up front if you do this it'll save you a ton of money and another

00:11:45.540 pro tip plus plus is clod code remote and i use this every time i leave my laptop where in your

00:11:51.600 terminal window where you're writing code you can say forward slash remote control and it will

00:11:55.980 connect to your clod app on your phone it keeps writing code and i did this three or four times

00:12:00.800 while i'm mountain biking on my phone i'm writing code and that is how i got this because i took a

00:12:05.920 berm and my front tire washed out. Now we just talked about how you build tools, but what if we

00:12:11.000 could learn how to build team members? Level six, the agent orchestrator. The orchestrator is the

00:12:17.280 agent you design that actually runs something. So now it's a loop that keeps running a department,

00:12:23.320 a process, a workflow, and you're no longer involved because you took all these pieces we

00:12:27.960 just talked about and you plugged it into the Claude agent. Now you're the human on the loop,

00:12:33.340 not the human in the loop.

00:12:35.180 Claude stops being just the tool

00:12:37.000 and it now becomes infrastructure.

00:12:39.040 Now, there's a million ways to build an agent,

00:12:40.900 but instead of telling you how to build just one,

00:12:42.860 I wanna give you a framework

00:12:44.220 to teach you how to think about building agents.

00:12:47.040 First, you have to start with one main agent.

00:12:49.940 This is your agent.

00:12:51.180 It could be your chief of staff agent.

00:12:52.600 It could be your admin agent.

00:12:53.620 I've got a CEO agent.

00:12:55.520 This is your orchestrator.

00:12:57.260 Mine is called Kai.

00:12:58.620 He doesn't do anything, really.

00:13:01.080 he just directs the other agents. I know this is crazy. Next, you have to go and create specialized

00:13:06.560 sub-agents. And each sub-agent owns a workflow. Your main agent communicates with the sub-agents

00:13:13.800 and tells them what to do. The third is you connect your telegram to talk to your agent

00:13:18.060 through your phone. Reese is my real estate agent. He finds deals of investments that I can do. And

00:13:24.040 Kai checks in with Reese every day and reports back to me because he does it with all my agents.

00:13:28.840 All done without me having to do anything.

00:13:31.360 The agents think, they decide,

00:13:33.640 they execute within the parameters.

00:13:35.780 See, we're building the machine that runs the machine.

00:13:38.480 Your main agent is that machine.

00:13:40.540 Now, I use my own platform called Apex,

00:13:42.980 so you can go check it out at apex.host,

00:13:44.580 but you can use Claude to do this.

00:13:46.420 Actually, behind Apex, we use Claude as the main agent model.

00:13:50.420 Apex just makes it more secure

00:13:51.880 and way easier to interact with it.

00:13:53.880 Now, here's a pro tip.

00:13:55.080 I have a critique agent that my main agent will ask

00:13:57.760 to go and review things like copywriting

00:13:59.800 or research or anything so that it's done right,

00:14:02.340 it creates a list of notes to make it better,

00:14:04.280 gives it back to the agent, it runs it again

00:14:06.520 so I always get the best output.

00:14:08.200 So now do you see how I'm no longer involved in the work?

00:14:11.580 I'm the human on the loop.

00:14:13.660 My partnership with the AI allows me to do way more

00:14:16.980 than if I was having to sit there

00:14:18.320 and process information through a pipeline.

00:14:20.560 And you can do this too.

00:14:21.960 Most people are just collecting cloud features.

00:14:24.100 They're like, oh, I use that, I use this,

00:14:26.100 but they don't actually make it part of their habits.

00:14:28.720 So here's what I need you to do.

00:14:30.040 Choose one of these features

00:14:31.640 and make a commitment to do 30 days in a row of using it.

00:14:35.860 Maybe it's a cloud browser extension.

00:14:37.200 Maybe it's advanced and say,

00:14:38.460 okay, every day I'm gonna work on an agent.

00:14:39.960 I don't know what it is for you.

00:14:41.220 Leave a comment below

00:14:42.140 and let me know what you're committed to.

00:14:43.580 And remember, if you want my AI company OS playbook,

00:14:46.560 the same one I use in all of my businesses,

00:14:48.620 just DM me YouTube OS on Instagram

00:14:50.500 and I'll send it right over.

00:14:51.740 And if you wanna know the six most profitable

00:14:53.480 AI businesses to start,

00:14:54.980 click here and I'll see you on the other side.

Dan Martell - June 15, 2026

Full Claude Guide： Beginner to Pro in Under 15 Minutes

Episode Stats

Summary

Transcript