OpenClaw Full Tutorial for Beginners: How to Setup Your First AI Agent (ClawdBot)

In this video I walk you through a complete OpenClaw (ClawdBot) install from scratch — covering deployment on a dedicated device, Telegram bot setup, Open Router API keys, security hardening, and Brave web search integration.

*https://www.youtube.com/watch?v=BoC5MY_7aDk
**https://300.ya.ru/v_TnnelUiG

пересказ видео

00:00:00 Введение в Open Claw

  • Объяснение, как установить Open Claw безопасным способом.
  • Советы по использованию в первые дни.
  • Проблемы, с которыми сталкиваются пользователи: страх перед установкой или чрезмерное волнение.

00:00:32 Подключение к Telegram и безопасность

  • Подключение Open Claw к Telegram.
  • Рекомендации по безопасности развёртывания.
  • Настройка API-ключа для использования моделей ИИ.

00:01:22 Выбор устройства для установки

  • Не устанавливать на основной компьютер с секретной информацией.
  • Рекомендации по использованию старых устройств, Mac Mini, Raspberry Pi или VPS.
  • Предупреждение о рисках использования сторонних изображений в один клик.

00:02:28 Установка на Raspberry Pi

  • Использование Raspberry Pi для демонстрации установки.
  • Команда для проверки и установки Open Claw.
  • Подход MVP: ограничить настройки и постепенно улучшать конфигурацию.

00:03:22 Безопасность и Open Router

  • Важность тщательного планирования безопасности.
  • Рекомендация использовать Open Router для доступа к моделям ИИ.
  • Преимущества Open Router: один ключ API для всех моделей.

00:04:36 Настройка Open Router

  • Создание учётной записи в Open Router.
  • Добавление кредитов и создание API-ключа.
  • Генерация токена и настройка шлюза.

00:05:27 Создание Telegram-бота

  • Создание нового бота в Telegram.
  • Ввод имени бота и настройка политик доступа.
  • Использование списка разрешений для управления доступом.

00:07:00 Завершение установки

  • Проверка работы панели управления Open Claw.
  • Отправка сообщения для тестирования работы агента.
  • Важность первого разговора с агентом для понимания его возможностей.

00:07:21 Работа с агентами

  • Возможность развёртывания разных агентов для разных задач.
  • Первый диалог с агентом для определения целей и задач.
  • Обращение агента к своим файлам MD для получения информации.

00:08:19 Введение в MD-файлы агентов

  • MD-файлы содержат инструкции для агентов.
  • Bootstrap MD используется при первом запуске агента для определения его личности и целей.
  • Агент заполняет поля на основе предоставленной информации.

00:09:01 Настройка имени пользователя и проектов

  • Имя пользователя MD определяет, кем вы являетесь для агента.
  • Можно указать часовой пояс, особые указания и информацию о проектах.
  • Агенты могут редактировать свои файлы по запросу.

00:09:56 Общение с агентом через Telegram

  • Использование Telegram для общения с агентом.
  • Преимущества Telegram: доступность и удобство.

00:10:24 Проверка безопасности установки

  • Агент выполняет проверку безопасности установки Open Claw.
  • Удаление конфиденциальной информации из утечек и логов.
  • Усиление безопасности путём выполнения рекомендаций агента.

00:12:09 Настройка веб-поиска

  • Установка API-ключа Brave для веб-поиска.
  • Настройка веб-поиска через терминал.
  • Проверка работы веб-поиска с помощью агента.

00:13:14 Автоматизация задач

  • Начало с небольших задач, например, получение новостей об искусственном интеллекте.
  • Оптимизация задач по мере необходимости.
  • Использование MD-файлов для организации задач.

00:14:41 Создание команд агентов

  • Создание отдельных MD-файлов для различных задач.
  • Подключение новых агентов на основе существующих файлов.
  • Планирование будущих видео о создании команд агентов и командного центра.

In this video

Chapters
Transcript
Search transcript
Chapter 1: Introduction and setup
0:00
0 seconds
Hello legends. In this video, I’m going to show you how to install OpenC Claw from scratch. I’m going to show you how to deploy it in a safe way. And I’m also going to give you a bunch of tips and
0:08
8 seconds
tricks on how to use it for the very first few days so you can get the most value out of it. I’m finding that a lot of people fall into two buckets. One, they’re either too scared to or they
0:16
16 seconds
don’t know how to install OpenClaw. And two, they do install OpenClaw, but they get too excited. They try to do too many things with it all at once. And then
0:23
23 seconds
they ultimately waste a lot of time, a lot of tokens, and then they lose their motivation, and they stop using it. So, in this video, I’m going to take you through a brand new install of OpenClaw.
0:32
32 seconds
I’m going to show you how to connect it up to your Telegram. I’m going to give you a couple of security prompts that will help you secure your deployment even better. The OpenClaw agent is
0:41
41 seconds
actually able to scan a device that it’s on. Understand how you deployed and make some recommendations on how to just harden that deployment. I’ll show you
0:48
48 seconds
how to set up your API key so you can use different AI models. I’ll show you how to set up your web search. I’ll give you an overview of all the different MD files and how the very first
0:56
56 seconds
conversation works when you first deploy your agent. And then finally, I’ll give you a couple of tips and tricks on how best to use OpenCore for the very first
1:04
1 minute, 4 seconds
few days. So instead of trying to do absolutely everything all at once, I recommend the scaled back approach where you spend the first couple of days just getting a feel for things and then brain
1:12
1 minute, 12 seconds
dumping all of your goals, your challenges, whatever tasks, whatever things you’re working on to your OpenClaw agent and then asking it to help set up the very next steps for you.
1:20
1 minute, 20 seconds
So OpenClaw is pretty straightforward to install. You don’t want to install it on your main computer where you’re logged into all of your accounts, you have any secret information like files, documents
1:29
1 minute, 29 seconds
or passwords because if your open claw does get compromised, then it can access all those files on a computer. So really what you want to do is get a dedicated
1:36
1 minute, 36 seconds
device and that could be an old laptop or a computer you have lying around. It could be a Mac Mini, it could be a Raspberry Pi, which is what I use, or it
1:45
1 minute, 45 seconds
could be a VPS, so a virtual server. If you’re using a virtual server, OpenClaw doesn’t recommend using a third-party one-click image. So, a one-click
1:53
1 minute, 53 seconds
deployment is something like Digital Oceans button over here where you can click this and then it spools up a custom configured version of OpenClaw
2:00
2 minutes
directly on a server. And same thing with Hostinger. You just click on this one-click deploy button and then you have a deployed OpenClaw on their server. So, the upside of using a
2:09
2 minutes, 9 seconds
one-click image is that it’s very easy to get started. But the downside is that over time you might want to change some of the configurations. And if you’re
2:17
2 minutes, 17 seconds
using someone else’s preconfigured version of OpenClaw, it might just be difficult or maybe impossible to change some settings later on. So, so the
2:24
2 minutes, 24 seconds
recommendation is just to start from a blank slate and then do it yourself. So, I’m going to be using a Raspberry Pi for the demo today. And to get started, you just need to copy this command and then
Chapter 2: Installation process
2:33
2 minutes, 33 seconds
open up a terminal on a device that you want to install OpenClaw. So, for me, I’m on my Raspberry Pi. Okay, so I’m going to open up my Raspberry Pi, open up a terminal, and then paste in this
2:41
2 minutes, 41 seconds
command. This command is going to do absolutely everything for you. It’ll first look at your device and make sure you got all the prerequisites, and if you don’t, it’ll just download them.
2:49
2 minutes, 49 seconds
Second, it’ll download and install OpenClaw, and then third, it’ll finalize the installment, and then it boots up the onboarding process. Now, a
2:56
2 minutes, 56 seconds
suggestion here is that I would not get bogged down on all the different settings that you can choose. I would just do the MVP approach. Just get the bare minimum in place. Use OpenClaw for
3:05
3 minutes, 5 seconds
a couple of hours, maybe a day. Begin to understand exactly where you want to use it, how you want to use it, what do you want it to do for you, and then when you
3:13
3 minutes, 13 seconds
know uh more about what you want to do, wipe OpenClaw from your computer, restart the onboarding process, and then be even more deliberate and specific with the configuration and security
3:21
3 minutes, 21 seconds
settings. OpenClaw even mentions in their security document that there’s no perfectly secure setup, and the goal is to be deliberate about who can talk to
3:29
3 minutes, 29 seconds
your bot, where the bot is allowed to act, and what the bot can touch. So only after you use openclaw for the first time and you begin to understand its
3:37
3 minutes, 37 seconds
magic. Only then can you actually know what you want to do with that and then be very deliberate with the options you choose. So let’s keep going and say yes
Chapter 3: Model configuration
3:44
3 minutes, 44 seconds
we understand the security risks and let’s opt for manual onboarding. Let’s keep local gateway accept this directory. And for here my
3:53
3 minutes, 53 seconds
recommendation at least for the start is to use something called open router. And if you haven’t heard of Open Router, this is a provider that gives you one single API key and then access to pretty
4:02
4 minutes, 2 seconds
much every single AI model that you’d ever want to use. You can access all the OpenAI models, all the Google models, all the anthropic models, and then you get access to these open source models.
4:11
4 minutes, 11 seconds
And the benefit of using these open- source models that are very cheap to run. It’s 30 cents per million input tokens and $110 per million output
4:19
4 minutes, 19 seconds
tokens. And this is said to be on par with models like Claude called Opus 4.6.
4:23
4 minutes, 23 seconds
And the comparison here is that for Opus 4.6, 6. It’s $5 per million input tokens and $25 per million output tokens. Now,
4:31
4 minutes, 31 seconds
there are some use cases that you might want to use Opus 4.6, but the Miniax, I’ve been experimenting with it and it’s actually it’s very good to get started
4:38
4 minutes, 38 seconds
with and be your daily driver. So, then the value is you get one single API key from Open Router and you can access every single one of these providers and
4:46
4 minutes, 46 seconds
all of the different models. Let’s hit enter for Open Router. After you create an account from Open Router, go into the credits tab, add about 5 or 10 bucks
4:54
4 minutes, 54 seconds
just so you can get started. Then go across to API keys. Create a new API key. Give it a name and click create.
5:00
5 minutes
And then copy that key. Paste it into here and hit enter. And then to get us started, I’m just going to go down to Claudson 4.6. Let’s hit enter. And then
5:08
5 minutes, 8 seconds
hit enter and accept this port. In my case, I’m already running another instance of OpenClaw that is attached to this port. So I’ve got to change mine.
5:15
5 minutes, 15 seconds
Hit enter. We’re going to keep the gateway bind on loop back. We’ll go with the recommended token. Leave tail scale off and then hit enter to generate a
5:23
5 minutes, 23 seconds
token. and let’s configure our first channel. We’re going to go with Telegram bot. Once you’ve signed into your account for Telegram, just start a new conversation with the bot father. Then
Chapter 4: Telegram integration
5:32
5 minutes, 32 seconds
in your conversation, go forward slash and newbot and hit enter. Enter in a descriptive name for the bot. And then for the second name, just make sure you
5:40
5 minutes, 40 seconds
end in bot. And you cannot use the dashes. You have to use underscores if you want these kind of spaces. Let’s hit enter. The bot’s been created. So we can
5:47
5 minutes, 47 seconds
now copy our API key. Paste the bot token into here. Let’s go all the way down to finished and hit enter. and let’s configure the DM access policies
5:55
5 minutes, 55 seconds
now. So just like in the install instructions, you want to be very specific where you deploy your bot and who can speak with your bot. My preference here is to use an allow list.
6:03
6 minutes, 3 seconds
For the allow list, you just get the user ID or the username and then you enter it into here and only that user can speak with your bot over time. If you want to add someone else from your
6:11
6 minutes, 11 seconds
team to be able to chat with the bot as well, you can just update the allow list. Back in Telegram, you just want to get your username or user ID. So let’s
Chapter 5: Finalizing onboarding
6:19
6 minutes, 19 seconds
click on this hamburger menu. Click on your profile. And I find that sometimes copying the username doesn’t actually work in this setup. So if you go into the URL of this page, you’ll actually
6:27
6 minutes, 27 seconds
find your user ID. Only copy the digits and nothing else. Paste it into here and hit enter. I’d recommend to start off by not configuring any skills, but just to
6:35
6 minutes, 35 seconds
take a look. And here you have a bunch of pre-built tools that can plug into your OpenClaw agent. I’m just going to skip this for now. So I’m going to hit spacebar to select skip and then hit
6:42
6 minutes, 42 seconds
enter. I’m not going to supply any more API keys. So I’m going to go no, no again and again and then again all the way through to the end. And over here,
6:51
6 minutes, 51 seconds
let’s just enable all of these different hooks. Once we’ve done, let’s hit enter and click yes and hit enter to install the gateway service and then yes for node. Let’s hit enter to enable bash.
7:00
7 minutes
And with that, we complete the onboarding. So at this stage, we have openclaw set up and installed on our machine. Up in the top section over here, you can see the dashboard is
7:08
7 minutes, 8 seconds
ready. Let’s just copy the URL link with the token. Copy that and then paste it into your terminal. And here you have your openclaw dashboard. Let’s test if
Chapter 6: Initial conversation
7:17
7 minutes, 17 seconds
this works by sending in a message. Hi there. And here’s our response. Now, this conversation is very important. As
7:24
7 minutes, 24 seconds
you can see, we’re basically asked to help the agent understand who it is, who we are, and what kind of goals we have.
7:30
7 minutes, 30 seconds
Now, every time you deploy a new agent for OpenClaw, so right now, this is one agent we have, but we can technically go into the agents tab and then just deploy
7:38
7 minutes, 38 seconds
another agent for research, another agent for writing, whatever, whatever use case you need. Every time you deploy a new agent and have your very first
7:47
7 minutes, 47 seconds
conversation, they’re always going to have a similar experience like this, asking who you are, who they are, and what their purpose is. So, at this stage, you want to spend 5 or 10 minutes
7:55
7 minutes, 55 seconds
just initially having this conversation with the agent. If you do what I said before where I’m like, just install it first, begin to understand what it’s capable of doing for you, and then you
8:04
8 minutes, 4 seconds
begin to understand what you want to do with it. Same thing goes for this conversation. Just get the first conversation out of the way. The important thing is that the agent actually goes across to its own files.
8:13
8 minutes, 13 seconds
So over in the agents tab, if we click on the files over here, we got a bunch of MD files. So this is just marked down. It’s just a format of a file like
8:21
8 minutes, 21 seconds
a doc or a PDF, but in this case, it’s just MD. So some of the files that we have here are the agents MD file is just the instruction set for the agent
8:30
8 minutes, 30 seconds
whenever it’s having a conversation with you. The very first run, it’s told to actually use the bootstrap MD. So the bootstrap MD is just that first
8:38
8 minutes, 38 seconds
onboarding conversation where it’s been prompted to say, «Hey, figure out who you are, who you’re working with, what the goals are.» Um, and to basically
8:46
8 minutes, 46 seconds
help build some personality and character for this agent. You have the solemn MD, which is like the personality character. You have the identity MD,
8:54
8 minutes, 54 seconds
which is the agent’s identity. So once you tell it exactly who it is, how you want it to operate, it’ll fill this out.
9:01
9 minutes, 1 second
And then you have the user MD. The user MD is essentially who you are to the agent. So you just give it your name. Uh
9:08
9 minutes, 8 seconds
you can give it your uh time zone that you’re in, any special notes about communicating with you, and some more context around projects that you want to create. Over time, uh you’ll actually
9:17
9 minutes, 17 seconds
find that you’ll evolve whatever files you have here. If you create a research agent, you might tell create a research.md file for you and just start
9:25
9 minutes, 25 seconds
putting in different bits and pieces of research that you’re working on. Now these agents are able to edit those files whenever you want or they want. So
9:33
9 minutes, 33 seconds
after you have this first conversation, if you find in two days time that you want to update like how the agent actually communicates with you, which is definitely reasonable, you can tell the
9:41
9 minutes, 41 seconds
agent to go back and just update its own configuration file. So I just finished my initial conversation with OpenClaw. I defined who I was, who the agent was,
Chapter 7: Securing the setup
9:49
9 minutes, 49 seconds
and what our goals were. And then I reset our session so we can start fresh.
9:53
9 minutes, 53 seconds
Let’s now test out the Telegram connection to make sure it works. I’m going to go back to our botfather and open up the chat. Since it’s the first conversation we’re having, let’s go and
10:01
10 minutes, 1 second
start the combo. And here we have the message from our agent. Now, my preference is to just use Telegram for all the communication. I don’t really like using the dashboard to send and
10:09
10 minutes, 9 seconds
receive messages. Plus, it’s on my phone. And literally, wherever you are, you can just access and speak with your agent. Now, the very next thing you want to do before actually going away and
10:17
10 minutes, 17 seconds
giving your agent tasks and starting to see exactly what’s possible, I recommend you run a couple of these prompts just to harden security of your setup. The
10:25
10 minutes, 25 seconds
first prompt is basically saying, «Hey, can you please run a security audit of our setup, so of our device and our deployment of OpenClaw, and let me know
10:33
10 minutes, 33 seconds
if we have any issues that we should fix.» And since the agent knows how OpenClaw was built and has access to different commands for OpenClaw, like security audits or health checks, it’s
10:42
10 minutes, 42 seconds
able to look at your device and then recommend some changes. I recommend you actually read the response and understand what the agent’s talking about. And if you agree with everything it said, you can just reply back and
10:51
10 minutes, 51 seconds
say, «Can you please fix everything?» The next command you want to run is around redacting sensitive information from leaking in logs. This is another
10:58
10 minutes, 58 seconds
security feature that you can do in the back end of OpenClaw. There’s actually a specific feature where when you’re having a conversation, sometimes the metadata of the conversation might have
11:07
11 minutes, 7 seconds
an API key or a specific user ID that you don’t want to leak out into the wild. So, we’re just asking OpenClaw to enable that feature and just make sure
11:15
11 minutes, 15 seconds
that we’re redacting that sensitive information. And this is another really good scan to do. As you can see from default, some of these tokens were shared as plain text. So then the agent
11:24
11 minutes, 24 seconds
went through here and then it made some recommendations on how we can beef up our security and then including down here as well. So you can just reply back to the agent and say, «Yes, I agree with
11:32
11 minutes, 32 seconds
all your recommendations. Let’s do it.» Then the agent will go away and then fix everything for you. So at this stage, I’d say that you’ve doubled the security of your instance just by sending those
11:41
11 minutes, 41 seconds
two very basic prompts. Now we have some breathing room and can start actually figuring out what this agent can do for us. This first part is going to be a relatively personal experience where
Chapter 8: Web search integration
11:50
11 minutes, 50 seconds
you’re going to be prompting up the agent based on your specific tasks and goals, but I’m going to give you a couple of tips and pointers on what you could possibly do. So, the first thing is we want to give our agent the ability
11:58
11 minutes, 58 seconds
to search the web. So, we’re going to install the Brave API key. The two default options are using Brave Search or Perplexity for Brave Search, which is what I’m using right now. You get a
12:07
12 minutes, 7 seconds
bunch of free API credits that you can use. I definitely recommend starting with this or since we’re using open router already to run our claude sonnet
12:15
12 minutes, 15 seconds
model. You can use open router to use the perplexity web search as well. Let’s go across to Brave and get our API key.
12:20
12 minutes, 20 seconds
After you create an account, you don’t have to upload any credits. Right now on this plan, you get one request per second, which is actually pretty good.
12:27
12 minutes, 27 seconds
You can just let your agent know to slow it down its internet requests and you get 2,000 requests per month total. So, this is quite a high capacity. Let’s
12:35
12 minutes, 35 seconds
click on add an API key. Give it a name and hit add. Let’s copy that key. And we can either go back into the agent conversation, give it the API key and
12:43
12 minutes, 43 seconds
say, «Hey, can you please set up Brave web search?» Or we can go with these instructions and copy this command. Back into your terminal, you can keep going
12:51
12 minutes, 51 seconds
on from the previous session you had. Or even if you start a new terminal session, just hit enter. Let’s leave it as local and let’s enable web search for
13:00
13 minutes
Brave. Just get that API key one more time. Paste it into here and hit enter.
13:04
13 minutes, 4 seconds
Let’s enable web fetch. And it’s complete. Now, let’s just make sure it works. So, I’m going to get the agent to hit the Brave API and let us know if it
13:13
13 minutes, 13 seconds
was successful or not. Awesome. So, now we have web search. So, let’s set up our first daily task. So, I’m asking the agent in plain English, hey, can every
Chapter 9: Automating daily tasks
13:21
13 minutes, 21 seconds
day at 8 a.m. you send me the latest AI news? Now, a tip here is that I would just start with something. Start with something small. And typically, I will
13:29
13 minutes, 29 seconds
run this automation or a task or whatever I’m doing for a couple of rounds. So, maybe like a couple of days.
13:35
13 minutes, 35 seconds
And as you go on, you can just get the agent and say, «Hey, you got me the AI news. Can you also let me know the weather or can you actually give me a
13:42
13 minutes, 42 seconds
research report or can you give me some YouTube video ideas?» So, as you go through, you can just tell the agent to optimize whatever task it’s doing. Then you can go back and change the scope of
13:51
13 minutes, 51 seconds
work by itself. Now, the final tip that I want to give you is the use of MD files, markdown files. So, for the first few days that you use OpenClaw, it’s
Chapter 10: Tips for usage
13:59
13 minutes, 59 seconds
going to be super exciting and most likely you’re going to try and do absolutely everything all at once. A very common thing that people do is, hey, I’ve got one agent. Why don’t I
14:07
14 minutes, 7 seconds
just deploy another two or three agents and you start getting them to do tasks for me. The difficult thing about that is that you don’t fully know exactly
14:14
14 minutes, 14 seconds
what you can do with OpenClaw. Instead of like wasting tokens, wasting your time doing unnecessary things, for the first few days, I recommend just sending
14:22
14 minutes, 22 seconds
a bunch of voice notes across to your OpenClaw and explaining here’s the things that I’m thinking about. Here’s what’s on my mind. Here’s a couple of goals that I want to achieve. And then
14:30
14 minutes, 30 seconds
piece by piece get your open claw agent to start building out separate MD files.
14:35
14 minutes, 35 seconds
For example, a research.m MD and then to put the research tasks that you mentioned from your conversations or travel.m MD. And if you ever speak about
14:43
14 minutes, 43 seconds
wanting to travel overseas or do things locally, it’s going to start putting those tasks and placing them in those MD files. So then over a couple of days,
14:50
14 minutes, 50 seconds
you’ll naturally build out two or three really solid MD files with a bunch of tasks immediately embedded into them. So then you can come back to your primary
14:58
14 minutes, 58 seconds
agent and just say hey based on the MD files that we have can you spin up multiple agents and then it’ll be very easy for you to configure those agents
15:06
15 minutes, 6 seconds
and then give them tasks so they can hit the ground running. Now my next steps are to be building out a bit more of a series about openclaw. So my next video
15:13
15 minutes, 13 seconds
will be about building an agent team and also building out command center where you can actually communicate with multiple team members on different tasks and projects. If you want to stay up to
15:21
15 minutes, 21 seconds
date with that video, please subscribe to my channel, drop a like on this video, or drop a comment saying that you’re excited to see the next one.
Sync to video time

Поделиться: