How to Set Up and Use OpenClaw: The Ultimate AI Agent Tutorial

My name is Charlie and this is a full comprehensive tutorial on how to use Open Claw. Now, it's going to be a long video, but it's also going to be the only video you need to watch on Open Claw. And if your goal is to set up your own Open Claw today, well, you are in the right place because in this video, I'm going to teach you exactly how to set up and use Open Claw. It's

honestly one of the coolest AI tools I've ever come across. I use it within my own businesses and the best part is that Open Claw is completely free. It's an open-source AI agent that connects to your messaging apps like Telegram, WhatsApp, Signal, Discord, whatever you want to use. We'll talk more about what I use it for in my own businesses, but essentially we're talking like automating tasks, managing files, sending emails, controlling actual applications, and so much

more, right? It's pretty much like having your own super smart assistant that lives inside your phone. All you need to do is message it just like a person asking it to do something and it's going to actually do it. So, if you guys follow along step-by-step with this video, which is what I recommend doing, you're going to have your own Open Claw agent fully set up and running by the time this video is over. Actually, the

initial setup will only take about 15 to 20 minutes, but then I do recommend playing around with it and sticking around for the second half of this video, too, where I basically take you guys through some actual use cases and how you guys can use Open Claw to help run your business or your life. And don't worry, you guys, if you're not super technical, I'm not super technical, you don't need to worry because I'm going to

walk you through the entire thing and it's actually not too bad at all. So, in the first part, we'll talk about what Open Claw actually is and why you should have one. And then I'm going to walk you through the entire setup, right? So, we'll go step-by-step from having zero Open Claw at all to a fully running claw on a server connected to Telegram, WhatsApp, and your Google Workspace. I'll also, of course, take you guys through

the whole setup, picking the LLMs you want to use and connecting them via API and a whole bunch of other really important, crucial things you need to do when setting up your Open Claw. I've watched a lot of videos on this topic. Many of these videos miss out on a lot of these things, especially the security portion of it. So, yeah, this is going to be super comprehensive. It's really important that you guys actually go through

it step-by-step. And then, like I said earlier, in the second half, I'm going to basically show you what Open Claw can really actually do. Things like web scraping, working inside Slack, even generating videos. I really think it's going to completely change how you think about what an AI assistant can be. All the links and resources that I talk about in this video are going to be in the description down below. So, make sure to reference it.

And yeah, let's get started with the video. So, what is Open Claw on a super high level, right? So, Open Claw, it's a free, open-source AI agent. This means that you don't need to pay for the actual software. It's completely free. There are some running costs, but we'll definitely cover that. So, unlike ChatGPT or Claude, where you're just chatting back and forth, Open Claw can actually go and do a lot of things for you. So, it

can manage your files, it can actually send messages, it can draft up emails, it can actually send these emails, you can connect to a bunch of different APIs to control other platforms and software, you can create workflows and actually automate them. And so, it's not just about like asking questions and brainstorming, it's about actually taking action. It is literally an assistant that is completely virtual, living inside your computer or VPS, and it's able to run 24/7.

Now, the way that you control and talk to your Open Claw is by actually messaging it through different apps. So, you can connect Telegram WhatsApp Signal Discord any of these apps. So, yeah, literally right now I can take out my phone, talk to my Open Claw, tell it to do something, and it's going to do it. And again, you are completely free to choose what LLM you want to use within Open Claw. You can connect Claude,

ChatGPT, Gemini, whatever you want. There's a bunch of other free ones as well, as well as really cheap ones, and this can really vary the costs of having an Open Claw. Because yeah, the main costs are going to be the computer or VPS that you host it on, plus the actual API costs, right? Using these LLMs cost money. But I'll show you guys how to do it properly and how to actually lower your costs in this

video. Okay, so now let's talk about where you're going to run your Open Claw. So, you're probably wondering, how can I actually run this thing? Like, what equipment do I need to use? Now, the first option is you can run Open Claw on your own computer. But for me, I don't recommend doing that. Your laptop or your desktop, whatever, it goes to sleep, your Wi-Fi might go out, and if that happens, then your claw just stops

working. So, if you guys want it to be available 24/7 and able to respond to Telegram messages even when you're asleep, even when you're traveling, then you're going to probably want to have it running on a server. And I know you guys have probably heard of the whole hype of buying these Mac minis and running your Open Claw on that. That's definitely better than running it on your own personal computer because, like I said, Open Claw

can mess with a lot of your files. Because the thing is with Open Claw is it has full access to your computer. So, I don't recommend running it on whatever main computer you use. If something goes bad, then it could cause a lot of havoc, right? But the thing with an actual computer like a Mac mini is there's a huge upfront cost. You're going to be spending probably $500 to a few thousand dollars buying this computer,

setting it up, and then you also need to have it run continuously. So, for me, I chose to host my Open Claw on a VPS. A VPS is a virtual private server. It's basically a computer that lives in the cloud. It's always on, it's always connected. It's only like $5 to $10 per month and in my opinion, it's the best way to run Open Claw if you actually want it to be reliable. I personally run mine

on my Hostinger VPS, but I have some other options for you down below because you really can't go wrong. Now, one reason why I like Hostinger is because the pricing is super low. They also make it very easy for you to set up your Open Claw. Not only do they have a Docker manager built right into the control panel, but they also have Open Claw as a one-click install. And that's going to save you guys a

ton of headache. Now, I'll have a couple links down below. And essentially, the first option is the all-in-one Open Claw bundle. So, this is essentially Open Claw already installed and it's like a shared version of a Hostinger VPS. So, it's not your own VPS, plus it's a little bit cheaper. But if you want your own VPS, you're going to want to click on that second link. It's going to take you to a page that looks kind

of like this. It might change depending on when you're watching this video, but this allows you to basically have your own VPS plan that can be able to deploy Open Claw on your own. Now, for this video, I'm going to take you through this way to do it. The reason for this is just because it's a little bit more of a better install. You have your own VPS, which in my opinion is a little bit better.

It just gives you a little bit more control. But of course, if you don't want to mess with your own VPS, just go with the already managed Open Claw. That's also going to work fine. Now, as you can see, there are different VPS plans. There's KVM 1, KVM 2, KVM 4, as well as KVM 8. For most of you guys watching this video, the KVM 1 plan is going to be good enough. As you can see,

there's one VCPU core, you get 4 GB of RAM, 50 GB of disk space, as well as 4 TB of bandwidth. On the KVM 2 plan, you get basically double that, two VCPU cores, 8 GB of RAM, 100 GB of storage. The thing is, with your Open Claw, if you're not doing anything too crazy, you're not running local LLM models like Ollama, then the KVM 1 plan should be good enough. The KVM 2 is also really

great. It's a little bit more expensive, so it's really up to you guys what to choose. Just for this video, I'll show you the KVM 1 plan. We'll click on deploy. So, first it's going to have you select the time period. Now, for all you guys watching this video, I'm going to recommend either 12 or 24 months. 12 months is going to be a little bit more expensive per month than the 24 month. But the best

thing about the 24 month is that it locks in this super low rate for a long time. However, the price difference per month is not too big. Just for this video, I'll go with 12 months. I think that's a good starting point. I definitely do not recommend going with 1 month as the pricing just doesn't make any sense. You also get a free domain if you choose 12 months or longer. Now, here's going to be a

couple different options you guys can select or unselect. For this video, I'm going to uncheck this one because I will show you guys how to actually set up your API keys. You don't need these nexsos.ai credits and there's a better way to do it in my opinion. And then, this one, instant web scraping, I actually recommend getting this because then it makes the setting up the web scraping a lot easier and it comes with 1,000 free

credits. For these ones, I don't think most of you guys will need this. Some of you might want daily auto backup if you want. It's only $3 more per month. But for this video, I'm just going to leave it as is. And then for your server location, you're going to want to just choose the one that's the fastest and likely that's going to be the one that is closest to where you live. So, I'm in the

US, so I'm going to leave this as is, but of course, select accordingly. Now, if you guys click on the link, you can see the coupon code is automatically inputted here, which is going to save you an additional 10%. If it's not working for some reason, just put in code CCM10 or Charlie Chang, either of those work, and it's going to take 10% off. Doesn't matter which one you guys use. So, as you can see, we

are getting this KVM 1 plan for 12 months and it's just about $75. So, way, way, way better in my opinion, way cheaper than getting your own Mac mini. For me, it just makes a lot of sense to have your Open Claw running on a VPS. We'll click on continue now. And now you're going to enter in your email address, create a password, and then enter in your payment details. It's a very straightforward registration process. Now,

once you're in the Hostinger panel, what you're going to do is go to Docker manager and you're going to see your Docker projects listed there. Now, depending on how you sign up for your VPS, the next few steps might be a little bit different. If you want to go fully non-technical, I would recommend just doing the one-click install. Hostinger is going to take you through all these setup steps, which is essentially the Open Claw onboarding. This

is what you're going to do if you already have a VPS or if you're setting it up manually. But like I said, if you're doing it with the one-click install, some of these steps may not apply. So, we're going to hit Docker manager, then find Open Claw in the catalog, and then we're going to deploy it. So, here it's going to ask you to fill in API keys for AI providers. And if you already have one

like OpenAI, Anthropic, or Gemini, then you can fill that in right now. I'll have some guides down below that will show you exactly how to set up these API keys if you want to do it this way. But for me, I prefer to use OpenRouter and not a lot of people talk about this. But essentially, it just gives me one API key and I can switch between models super easily. It's a little bit harder to set

up initially, but personally, I think it is worth it. Okay, so now you'll see it show up in your project list. It's going to say created at first, and then once it's done deploying, the status changes to running. Once it is running, you're going to see a port number next to it. That's how you access your Claude's web interface. So, what we're going to do is click on that and you'll land on the Open Claude welcome

page. It's going to ask you for your gateway token. And to get this, you can grab that from your Docker manager page where it says gateway token. Just copy, paste it in, and then hit login. Now, the first thing you'll see is a message saying your AI provider API keys are missing, and that is completely okay. Now, Hostinger's one-click setup is really good for getting Open Claude running fast, but it doesn't give you the option to

pick your model provider during install. So, what we're going to do is we're going to run the setup wizard from the terminal instead. And this actually works out better because it gives you way more control. So, what I'm going to be using is OpenRouter, and if you haven't heard of what this is, it's basically a single API that gives you access to every major AI model out there. Claude gives you GPT, Gemini, GLM, MiniMax, you name

it. By the time you're watching this video, there might be new ones out there, but these are the ones that are currently popular. So, yeah, you have one API key, one account, and then you can basically switch between models whenever you want. So, what we'll do is head over to the Hostinger VPS web terminal. This is just a command line into your server, and you can sort of think of it like SSH but right in your

browser. We're going to exec into the Open Claude Docker container. So, what you'll do is you'll type docker exec space {dash} IT followed by your container name. And you can grab that from the Docker manager page. Then open claw onboard. This drops you into the Open Claude setup wizard. I know it seems a little bit intimidating, but trust me, this will be worth it. Okay, the first thing it shows you is a security notice, and this

is super super important. Like I mentioned, Open Claude is really powerful, which means it can do a lot on your server. And we're fine since Open Claude is containerized and it lives on the VPS. So, we're going to say yes. The wizard then detects your existing configuration and asks what you want to do. So, we are going to update the setting, so choose update. Next, it asks for your model provider, and this is where we're going

to pick OpenRouter. So, select that from the list, and it's going to ask for your API key. So, now what you're going to want to do is go to openrouter.ai, create an account there if you don't have one, and then generate a new API key. Feel free to pause this video and do that. And this is super important, you guys, save this key somewhere you can access later on because once you close that window, you're not

going to see it again. It's basically a password, and the next thing you're going to want to do here is add some balance into your account so that you can actually use the models. So, I'd recommend loading at least like 20 to 50 dollars into your accounts. We'll go with 40. That's going to last you a good amount of time. So, once you do that, you're going to paste the key in, and then it sets your

default model. I'm going to go with Claude Sonnet 4.6 as my main model, but you guys can pick whatever you want. And the beauty with OpenRouter is that you can always change or switch later on without changing providers. The other option with setting up your API keys, you're going to get your OpenAI API key, and you'll have to do these each individually. It's a little bit annoying if you want to go that route, and you'll have

different accounts. You'll also need to set up billing for each of those. So, that's why I prefer OpenRouter. Okay, now it's going to ask about channels. This is how people talk to your Claude. This is how you talk to your Claude. So, there's options like Telegram, WhatsApp, Discord, Slack. And what I'll do is I'll skip these for now, and we're going to set those up manually in just a second. It also asks about web search and

a search provider. You can skip those for now, too. Same thing with hooks, just skip those. Then it gets to skills. So, here you're going to say yes, and specifically install GOG. So, that is the Google Workspace skill, and we're going to use it later to connect your Gmail, calendar, drive, all of that. And finally, it asks how you want to hatch your bot. So, what you'll do is choose do this later. We're going to do

it through the web dashboard instead, which is a much nicer experience. Now, let's talk about accessing the Open Claude dashboard. Now, your Open Claude instance is running on the VPS, but the dashboard isn't exposed to the internet by default. And that's a good thing. You probably don't want it publicly accessible. So, what we're going to do is we need to create a secure tunnel from your computer to the server. This might seem bit complicated, but it's

really important that you guys do this. Not everyone accesses their Open Claude this way, but for me, it is the most secure way to do it. So, open a terminal on your local machine and run an SSH command to forward the port. So, it's going to look something like this. SSH space {dash} N space {dash} L, and then this whole string of characters, right? So, in this case, 55779 is my port, and depending on your VPS,

it's going to be different. To find this number, you're going to go to the Docker manager and replace your port number here. You're going to follow that with your username and then your server address. So, this creates a tunnel, right? If this seems a little bit complicated, I'll have another doc down below to really go through the fine details on this. Now, when you go to localhost and then your port number in your browser, it connects

straight to your Open Claude instance on the VPS, but securely through SSH. Okay, so once you open that in your browser, you'll see the Open Claude control dashboard. This is the gateway dashboard page. It auto-fills the web socket URL and your gateway token. So, just hit connect, and you're going to land on a clean chat interface with a model selector at the top. So, this is where you meet your Claude for the first time. And yeah,

let's wake your Claude up. So, we're going to type wake up, my friend, and your Claude comes alive, and it starts with a completely blank slate. All right, there's no memory, there's no identity, there's nothing. So, it's going to ask you who you are and what you want to call it. So, I'm going to tell it that my name is Charlie, and I'm calling my Claude Sam. It's going to save that, write it into its memory,

and now it knows who it is and who you are. So, every future conversation, it's going to remember. Then I'm going to give it context about who I am and what I need. So, I run YouTube channels, I care about efficiency, I want it to help with scripts, emails, research, keeping me organized. It's going to save all this, and now Sam knows how to be useful to me specifically. This is what makes Open Claude different from

a regular chatbot. It builds a really persistent understanding of you over time. I definitely recommend actually answering these questions. Don't give one-word answers because then it's not going to feel as personal. Okay, now let's connect your Open Claude to Telegram because this is probably the most popular way that people like talking to their Open Claude. So, first, make sure you have Telegram open, and if you don't, just pause this video and sign up. Or if you

don't want to use Telegram and you want to use WhatsApp instead, then you can skip to this time code later on. Okay, so we're going to open up Telegram. We're going to search for BotFather. This is Telegram's official bot for creating bots. So, we're going to send it {slash} newbot. You're going to give your bot a name. So, I'm going to call mine Sam, and then a username that ends in {underscore}bot. Something like Sam_Charlie_bot. This doesn't

need to be a unique name, so choose something that's quite unique that other people can't find. And after we do this, BotFather is going to give us a token. And you're going to want to copy that. So, now back in your Open Claude chat, tell your Claude you want to set up Telegram. I'll walk you through the steps, but really, all you need to do is give it the bot token. And once you paste it in,

your Claude is going to run the plugin command to add the Telegram channel. The gateway is going to restart and then wait for a bit, and once it comes back online, your Telegram bot is connected. Now, you're going to open the bot in Telegram, hit start, and you'll see a pairing code. Give this pairing code to your Claude along with the bot username. It's going to approve the pairing. It's going to link your Telegram user ID.

It's really important that you guys do these steps for security, otherwise other people can actually talk to your Open Claude, and we definitely do [music] not want that. So, now you're going to be able to talk to your Claude from Telegram, which is probably going to be on your phone, on on your laptop, and it's going to respond right there in the chat. Okay, if you prefer to connect to WhatsApp instead, because I know this is

a very popular messaging app that's used across the world, then yeah, this is super easy. It just works a little bit differently. So, instead of a bot token, it uses a QR code. It's the same as when you link WhatsApp web to your phone, which if you guys use WhatsApp, you're probably familiar with that. So, back in your terminal, you're going to run the Open Claude channel setup for WhatsApp. It's going to generate a QR code

right in the terminal, and you're going to open WhatsApp on your phone, go to linked devices, and then scan the QR code. Once you do that, you're going to be fully linked. Then it's going to ask for your phone number. So, put that in so that Open Claude can allow list you as the owner. And once that is done, choose finished to exit the setup. Then run Open Claude gateway restart to apply the changes. Cool, so

now your Open Claude can send and receive WhatsApp messages. So, I'll show you. If I send it a test message on the server, it's going to pop up on my phone instantly. Now, how it looks is going to be a little bit different than Telegram. Essentially, all your messages and your Claude's messages are sent from your account. So, it looks like you're having a conversation with yourself. In my opinion, it looks a little bit weirder, so

for the rest of this demo, we're going to actually just use Telegram or the Open Claw web UI instead. Just know that this is how you connect WhatsApp if that is your preferred messaging app. Okay, so now let's give your claw access to your Google account because this is probably going to be one of the first things that most of you guys want to do. The GOG skill gives your claw access to your Gmail, your calendar,

Drive, Contacts, Sheets, and Docs. So, that means that you can say check my inbox or draft an email and your claw is just going to do it. Now, if you have valuable files that you don't want to risk being deleted, I'd actually recommend giving your claw your secondary Google account. But, I think for most people, it should be okay just to give it your primary account. So, first your claw checks if GOG is already installed and

it is because we installed skills during setup. But, it does need credentials to connect to your Google account. So, you're going to head to Google Cloud Console, create a new project. I'm going to call mine Open Claw, then go to APIs and Services and enable the any APIs that you need. So, I'm going to enable Gmail API, Google Drive API, Google Sheets API, Google Calendar API, as well as Contacts API. Just search for each one of

these and then hit enable. Next, you're going to set up the OAuth consent screen. So, go to Google Auth Platform, create the branding, fill in the app name and your contact email, and for the audience, set it to external and add yourself as a test user. That's your Gmail address. Then, what you're going to do is go to clients and create an OAuth client ID. Choose desktop app as the application type. This is going to give

you a client secret JSON file and what you're going to want to do is download that file. So, here is where it gets really cool. So, send that JSON file to your claw through Telegram. Just attach it as a document. Your claw is going to pick it up, save the credentials, and then start the authorization flow. So, it generates an OAuth URL. So, what we'll do is we'll paste that into your browser, sign in with your

Google account, approve permissions. Paste that callback URL back to your claw and that is going to complete the authentication. Now, we just covered a lot of stuff, so just feel free to pause the video or rewatch that section because this is really important. So, your claw confirms it's connected. So, we have Gmail, Calendar, Drive Contacts Docs Sheets everything. You can tell it to draft a reply, mark things as read, delete emails, all from Telegram. For example,

I asked Sam to draft an email to Brista telling her our meeting is delayed to tomorrow and then wrote a clean, professional email, showed me the draft, and waited for my approval before sending. So, now let's talk about switching models and some of my extra setup tips. So, currently as I am making this video, Anthropic's models are super, super good, right? Opus, Haiku, Sonnet, but their pricing is also, you know, really high. And that's why I

like to use Open Router because we can always switch to a cheaper model whenever we want. Some of the other open-source models like GLM or MiniMax, they might not be as good as some of these, you know, high-end ones like Opus in capability, but they also cost way less. And the thing is, you don't need to use the best models for everything. What I like to say is for things that require a lot of thinking, that's

when you want to use the frontier models, the ones that cost more money because frankly, they are better at thinking. But, when you actually want to carry out different actions, that's when you don't need these super intense models. You can use a cheaper model that can perform the action just as well and it's going to save you significantly on your API costs. Okay, so to switch, you're going to need to add more models to the model

picker. So, just run Open Claw configure from the terminal again. It's going to detect your existing configuration and let you update whatever you need. And I recommend adding GLM 5 and MiniMax 2.5 as cheaper alternatives. By the time you guys are watching this video, it might be a little bit different, but do some research, find out which other models you can use that are much cheaper and just select those. These are significantly less expensive than Claude

or GPT, but they still handle everyday tasks just fine. Now, to restart your Open Claw instance after making these changes, go back to Docker Manager in Hostinger, click the three dots next to your project, and hit restart. This takes a few seconds and then you're going to be back in. Now, you guys can switch models on the fly. So, if you're doing something simple and you don't want to burn through, you know, crazy API credits, just

type in {slash} model in your chat and pick a cheaper model. I switched to MiniMax for a quick task and it worked completely fine. You can always switch back when you want the more premier models. All right, so that's it for the setup. You basically now have a fully running AI assistant on VPS. It's running 24/7 in a virtual cloud. It's connected to Telegram and maybe WhatsApp with access to your entire Google Workspace. Now, you're able

to message it from your phone. You can ask it to, you know, write emails for you, send them, manage your calendar. But, honestly you guys, we're just getting started because now that your claw is up and running, let me show you what it can actually do. So, here's a problem that you're going to run into pretty quickly with any AI [music] assistant. You're going to ask it to look something up and it either can't do it

or it tries to and then gets blocked. The reason for this is because websites have a lot of anti-bot protection. There are lots of JavaScript-heavy pages that won't render, captchas, and yeah, all this stuff basically stops your AI from actually getting the data that you need. Now, if your assistant can't reliably access the web, it's stuck working with whatever it already knows and that's a huge, huge limitation. So, that's where Oxylabs AI Studio comes in. It's

basically a web intelligence layer for your claw. So, it's going to handle all the messy stuff like bypassing anti-bot measures, rendering JavaScript pages, solving captchas. And what it does is it gives your claw back really clean, structured data that it can actually read and work with. Now, I wanted to talk about this first because obviously, we want our claw to have access to the web. But, secondly, if you guys actually deployed Open Claw through Hostinger, you

already have 1,000 free Oxylabs [music] credits, so you don't need to create any extra accounts, you don't need any extra API keys, and there's not much setup, right? It just works out of the box. Let me show you guys what this looks like. So, I'm going to ask my claw search the web for the latest AI news today and give me a summary. Now, behind the scenes, Oxylabs AI search is doing the heavy lifting. It's finding

relevant pages, it's pulling the content, and it's handing it back to my claw as really clean data. So, a few seconds later, I have a nicely organized summary right in my chat. Yeah, my claw didn't just guess this data, it actually went out there into the actual web and got it. But, let's actually do something more useful. So, say you're running an Amazon store and you want to monitor competitor pricing. I tell my claw, go to

Amazon, find the top 10 wireless earbuds under $50, and give me a report with the product name, price, rating, and number of reviews for each one. So, this is where the AI scraper is going to kick in. So, it actually goes to amazon.com, handles all the anti-bot stuff that would normally block a scraper, and extracts the data that I asked for. So, a minute later, we have a clean pricing report right in my chat. You can

see the product names, prices, ratings, review counts, it's all laid out. There's no spreadsheet, there's no copy-pasting, there's no clicking through listings for an hour. And you could set this up as a cron job, so it runs every morning automatically. Obviously, you probably wouldn't do that for a report like this, but just apply this to any report that you need on a daily basis. Now, let me show you guys something else. So, I'm going to tell

my claw, go to this competitor's website, navigate to their pricing page, and extract their current plan prices. So, the browser agent opens the page, it clicks through the pricing section, reads the content, and then comes back with structured data. An normal scraper would probably fail at this normally because it's all dynamic, loaded with JavaScript, prices are hidden behind drop-downs, but Oxylabs can actually do it. So, here's another really cool thing you guys can set up that's

really good for monitoring. So, I can set up a daily cron job that tells my and message me on Telegram only when something actually changes. So, it's basically an automated competitor watchdog running in the background while I sleep. You don't have to configure a separate scraping service and you also don't need to install anything extra, right? Like it's already there when you deploy through Hostinger. Now, if you didn't go through Hostinger for your Open Claw, then

you're going to want to do this all manually. Okay, so the next use case, and this is going to be Slack, right? So, I run a YouTube channel with a team and essentially, all day we have editors asking about deadlines, sponsors reaching out, thumbnail feedback going back and forth. We have content ideas scattered across like 100 different threads and Slack is where all that stuff lives. And if you guys have a business, you probably also use

Slack and there's probably a lot of communication happening every single day. Basically, keeping up with it is like a part-time job and we don't want to waste our time doing that. So, your claw can actually handle a lot of that for you. With the Slack skill, your claw gets full access to your Slack workspace. So, it can read channels, it can send messages, it can search conversations. So, let me show you how we actually use this.

Okay, I'm going to tell my claw, check the video production channel and give me a summary of where each video stands. [music] So, it's going to do that and it reads through the channel and a few seconds later, it comes back with a status update. Which videos are in editing, which ones are waiting on thumbnails, as well as what feedback the editors left. So, if you ever leave for one or two days and you come back

and you have 200 unread messages and you're like, I'm not going to read all that. Well, this can actually help a lot with that. Now, I'm going to say, draft a message for sponsors. Let them know the integration for Brand X is scheduled for next Tuesday's upload and ask if they've sent over the talking points yet. So, it's going to take that, it's going to turn it into a professional message with the right tone, it's going

to show me the draft, right? Like it's not posting it yet, it's going to ask me to confirm first and I'm going to say, okay, that looks good. Send it." And I'm done. You don't even need to open Slack to go through your Slack messages anymore. Now, here's another use case, right? I'm going to tell my Claude, "Search Slack for every message about long-form launch from the last 2 weeks and tell me what's still unresolved." My

Claude digs through the conversations, it finds the relevant messages, and it gives me a clear picture of what's been decided, what's still open, who's waiting on what. And we can even combine this with cron jobs. And I don't think I mentioned what cron jobs are, but essentially they are scheduled things that happen on a daily or weekly or whatever basis. Super helpful for a lot of business owners, as well as even for your life in general.

So, you can have your Claude send you a daily production digest every single morning. Here's what happened across video production, sponsors, and content ideas yesterday. You start your day knowing exactly what needs your attention without opening Slack once. Now, one thing I do want to mention before we move on is whenever you finish a task with your Claude, it's really important that you guys start a new session. So, just type {slash} new. This keeps your context

window clean, so your Claude doesn't get confused by old conversations, and it saves you tokens as well. Every message you send includes the entire conversation history, so the longer a session drags on, the more tokens you're burning. So, that's why I encourage people to do fresh sessions often, because that's going to save you a good amount of money. Okay, next let's talk about Remotion. Uh this one might blow your mind a little bit, and it's really

cool because yes, Claude can actually make videos for you. It can generate animated graphics, it can create data visualizations, it can create promo clips all from a text description. So, Remotion is that lets you create videos using code. And when your Claude has the skill, it can actually write that code for you. So, you describe what you want, and it builds the video. Cool. So, let's demo this. I'm going to tell my Claude, "Create a 15-second

animated video showing our monthly revenue going from $10,000 to $50,000 with a bar chart that grows over time. Put our company name Academy in the corner." So, this is going to take a little bit longer because it's writing the code, it's rendering frames, and stitching all this together. So, you can see it churning away. And after a couple minutes, you basically have a clean animated video with a growing bar chart, smooth transitions, and of course our

name in the corner. So, here's another one. So, say you're running a YouTube channel and you need an intro. So, I'll tell my Claude, "Create a 5-second channel intro with the text Tech Weekly that fades in with a blue gradient background." So, same thing, it's going to write the code, it's going to render it, and it's going to give me a polished video clip. Of course, I can change it if I want. But yeah, as you

can see, we didn't have to use any fancy software. It literally just did this for us. The cool part is that because it's code-based, everything is tweakable, right? We can change anything we want. So, I can say, "Make the background red instead, or slow down the animation." And it's just going to update it and re-render it. You can sort of think of it as having like a motion graphic designer on call. So, yeah, those are some

pretty big heavy-hitter ways you guys can use your Open Claude in your business or your daily life. But what I do want to emphasize is that this is literally just scratching the surface of the skills you can do. I can't fit all this stuff on one YouTube video, but if you want to see more detailed guides on any of these skills or steps that I showed today, I have step-by-step walkthroughs for all of them inside my

free school community. Yes, as of right now, it is completely free. We might charge a tiny bit per month in the future, but as of right now, like I said, it is zero cost to you guys, and I'll have a link down below in the description. So, anyways, I hope you guys enjoyed the video. I know it was quite long, but honestly, this video could have been 5 to 10 hours long. There's so much other stuff

to cover about Open Claude, but I felt like I didn't want to include everything in this one video. I want to be more of like a beginner-friendly video that essentially helps you guys actually take action, get your Open Claude set up properly. I also wanted to walk you through some of the most helpful use cases that I use it for. Again, I'll have a link to hosting a VPS down below. I'll also include some other options

for you guys. This video isn't sponsored by any of them in particular, but these are just my recommendations. I do want to say if you guys use any of my links, it will help support the channel and keep videos like this completely free. So, I do want to say thank you so much in advance. I'll have time codes to every single portion of this video, so you guys can basically go through this, skip around if you

need. good luck on building your own Open Claude. Super, super exciting. And with the way that AI is moving these days, you need to be using AI for your life or your business. If you don't, you're going to get left behind, and that's why I'm really passionate about making videos like this. I think it's really important that everyone stays up to date on the newest AI advancements or tech, whatever it is. Hopefully this video helped you

guys set up your own Open Claude and actually encouraged you to use it. Cool. So, if you guys got any value from this video, make sure to hit that like button, subscribe if you want to see more content just like this. a ton of videos about personal finance, entrepreneurship, AI. And my whole goal is to help you guys live a financially successful life. Thank you so much for your time. I genuinely appreciate it, and I'll see

you in the next one. Peace. [music]