How to Master Claudebot: Setup, Customization, and Secure Automation on a Hostinger VPS

Feb 05, 2026

•

4 min read

YouTube video ID: 3GrG-dOmrLU

Source: YouTube video by Matthew Berman — Watch original video

PDF

Introduction

Claudebot is quickly becoming one of the most powerful personal AI assistants. It can chat through any platform (Telegram, WhatsApp, etc.) and connect to services like Gmail, Slack, Asana, and more. This article walks you through installing Claudebot on a Hostinger VPS, customizing its personality, managing models, creating skills, and keeping everything secure.

1. One‑Click Installation on Hostinger

Visit the Hostinger link (provided in the video description) and click Deploy.
Choose a subscription length (1, 12, or 24 months). The author selects 24 months.
Apply the coupon code MatthewB for a 10 % discount.
Complete payment and proceed to the configuration page.
Enter your API keys (Anthropic, OpenAI, Gemini, XAI, etc.) and click Deploy.
Open the terminal on the VPS and run openclaw onboarding to finish the setup.
Test the bot in Telegram – send hello and receive a friendly reply.

2. Core Configuration Files

soul.md – defines the bot’s personality, tone, and avatar.
skills/ – folder of repeatable processes (e.g., browsing, email, Twitter). Each skill is a set of steps the bot can execute.
tools/ – code snippets (e.g., fetch.js) that let skills interact with external services.
identity.md – controls how the bot presents itself (name, emojis, vibe).
memory/ – stores everything the bot learns about you; you can prune it manually.
heartbeat.md – runs periodic tasks (default every 30 minutes) and can be extended with cron jobs.

3. Model Selection Strategy

Primary model: Claude Sonnet 4.5 – cheap, reliable workhorse.
Fallback chain: Gemini 3 Flash → Opus 4.5 → OpenRouter → optional local models.
Switch models on the fly with natural language (switch to Sonnet 4.5) or the /model command.
Remember: the chosen model influences the bot’s personality and susceptibility to prompt injection.

4. Building Skills & Tools

Tell Claudebot the service you want to connect (e.g., Gmail, Drive, Slack). If a skill doesn’t exist, the bot writes one automatically.
Example: a fetch.js tool lets the bot pull data from Asana.
You can integrate advanced agents like Cursor for complex coding tasks; note that such agents have minimal personality.
Browse community‑created skills at clawhub.com. Always scan new skills for malicious code before using them.

5. Scheduling with Cron Jobs

Create one‑off or recurring tasks via natural language, e.g., remind me to drink water in 1 hour → creates a cron job.
Complex schedules are possible, such as a rotating recycling reminder that parses a picture of the schedule and notifies you every Sunday.

6. Using Telegram Groups for Parallel Workflows

Set up a dedicated group, add Claudebot as the only member, and make it an admin.
Enable the bot to reply to every message (not just mentions) so each group acts as an isolated topic.
Benefits:
Keeps conversations on‑topic.
Saves context‑window space.
Allows multiple simultaneous projects (video research, ebook drafting, content analysis, etc.).

7. Daily Self‑Audit

Schedule a daily review of core files (agents.mmd, memory.mmd, tools, soul, identity, heartbeat, etc.).
The bot suggests outdated info, conflicting rules, or undocumented workflows and asks for confirmation before applying changes.

8. Multimedia Capabilities

Generate images with Nano Banana (create an image of a lobster).
Add voice output via 11 Labs.
Drag‑and‑drop images or audio into Telegram for instant processing.

9. Security Best Practices

Store API keys only in an .env file and never commit it to Git.
Run openclaw security audit on the VPS; fix warnings with openclaw security audit -d -fix.
Keep the VPS isolated from your personal devices.
Treat all external data as dirty – it may contain prompt‑injection attacks (e.g., malicious email content).
Prefer stronger models (Opus 4.5) for tasks that handle sensitive data.
Use built‑in prompt‑injection detection, but verify critical actions manually.
Limit the number of third‑party skills you import; let Claudebot generate its own when possible.
Always ask the bot to propose a plan before it makes changes to files or integrations.

10. Real‑World Use Cases

Video Idea Pipeline: Drop a link in Telegram → bot researches via Brave API, checks Twitter trends via Grock API, and creates an ASA task for the team.
YouTube Analytics: Bot fetches recent video performance using YouTube Data & Analytics APIs and posts a summary to Telegram or Slack.
Meeting Prep: Every morning the bot scans Google Calendar, filters external attendees, gathers background info, and delivers a concise briefing.

Conclusion

Claudebot transforms a simple chat interface into a fully automated personal assistant. By installing it on a secure Hostinger VPS, customizing its personality, selecting the right models, and leveraging skills, tools, and cron jobs, you can automate everything from content research to daily meeting prep—while maintaining strong security hygiene.

Claudebot, when set up on a secure VPS and fine‑tuned with the right models and skills, becomes a versatile AI assistant that can automate personal and professional tasks safely and efficiently.

Frequently Asked Questions

Who is Matthew Berman on YouTube?

Matthew Berman is a YouTube channel that publishes videos on a range of topics. Browse more summaries from this channel below.

Does this page include the full transcript of the video?

Yes, the full transcript for this video is available on this page. Click 'Show transcript' in the sidebar to read it.

Helpful resources related to this video

If you want to practice or explore the concepts discussed in the video, these commonly used tools may help.

Raspberry Pi 4 Model B 8gb Recommended

A low‑cost, always‑on mini‑computer that can host Claudebot locally, offering full control and isolation for privacy‑focused users.

Amazon →

Amazon Echo Dot (5th Gen) With Clock

Adds voice interaction to your Claudebot ecosystem, allowing hands‑free commands and notifications throughout the home.

Amazon →

The Pragmatic Programmer: Your Journey To Mastery, 20th Anniversary Edition

Offers timeless software development best practices that improve the reliability and maintainability of custom Claudebot skills.

Amazon →

Links may be affiliate links. We only include resources that are genuinely relevant to the topic.

Summarize another video

Full Transcript YouTube

I started using Claudebot as an
experiment. It has quickly become one of
the most important pieces of technology
in my life. And if you haven't heard of
Claudebot, it has been going absolutely
viral. It is the most personal and
capable AI assistant that I have ever
seen. It allows you to get real tasks
done by connecting the different
services that you use like Gmail,
Telegram, Asauna, Slack. Everything can
be connected together and you just chat
with your personal assistant with your
Claudebot directly through whatever chat
interface that you want to use. I
personally choose Telegram. Cloudbot
learns about you and continuously gets
better. It is truly incredible. But a
lot of people still don't know how to
use it, and that's what I'm going to
show you today. By the end of this
video, you will become a Cladbot expert.
But before we can actually use
Claudebot, we have to get it set up. And
there are a number of different ways to
do that. Installing it on your local
machine, using external hardware, or
what we're going to be doing today,
using a virtual private server. And
we're going to be using Hostinger to do
it, who is also the sponsor of today's
video. So, huge shout out to them. And
it is basically one click to get it up
and running on Hostinger. So, the
benefits of using a VPS are super simple
to set up. It is always on and it is
secure because it is completely isolated
from any of your personal devices. And
Hostinger is offering 10% off their
already cheap VPS service with one-click
Cloudbot install. All you have to do is
use the code MatthewB and go to
hostinger.com/matthewb.
I'll drop everything in the description
below. All right. And getting it set up
with Hostinger is dead simple. You just
come to this page. I'll drop the link
down below. And all you have to do is
click deploy. On this page, you can
select your time duration. And this is
where you're actually going to purchase
the VPS. Right here, you can select 1
month, 12 month, 24 month. I'll start
with 24 months. And right over here
where it says have a coupon code. Yes,
we do. Matthew B. Apply. And there you
go. You get 10% off. Now continue. And
there you go. Payment done. Then it's
going to bring you to this configuration
page. You could put your API key from
Anthropic, OpenAI, Gemini, XAI, anything
else. And then click deploy. And then
you can connect everything else you need
by opening up terminal and going through
the open claw onboarding. And of course
all of this is optional. And here you
can see we're good to go in Telegram. I
said hello and my clawbot says hey
what's up. All right. So the first thing
to really understand are all the
different MD files that you have that
really form what this Cloudbot is for
you. And there are a bunch heartbeat
identity memory soul tools and user. So
this is the soul.md file. This is
basically defining what the personality
of your Claudebot is going to be. Feel
free to come in here and edit as you see
fit. Then we have the skills. This is
really what makes Cladbot so incredibly
capable. This is where you give it all
the skills that help it browse the web,
check your email, access Twitter, and
all of your custom skills as well. So,
anything that you tell Cloudbot to do
that it doesn't already know how to do,
it's going to create a skill from that
and save it here. You can think of a
skill as a repeatable process. Anything
where you have multiple steps and you're
using different tools along the way. And
so that's defined here. So we have the
humanizer skill for example and a whole
bunch more in here. And the skills
reference tools. And tools are pieces of
code that allow it to actually
accomplish those things that the skills
want to do. So here's an example. Here's
fetch.js.
And it's basically a JS file that allows
my agent to plug into ASA really easily
and grab information from my ASA board.
And again, you don't need to write any
of this. It's all done through natural
language through Telegram or WhatsApp or
wherever you're using your Claudebot.
Then we have the identity file and it's
very similar to Soul in a lot of ways.
It basically tells Claudebot how it's
actually going to interact with you from
a tone and emojis perspective. So we see
name, Claude, vibe, emoji, avatar, and
so that's what's in the identity file.
Cloudbot now has a memory folder in
which it stores all of its memories
about you. And you can just let it store
everything. You can always go in there
and prune things that you don't think it
needs to know anymore. But for the most
part, there's no harm in just having it
store most of what it learns about you.
Then we have the heartbeat file, which
basically runs every so often. I think
by default it's every 30 minutes but
it's whatever you want to have happen on
a recurring basis but we also have cron
jobs and I'll get to that in a moment
and again this entire cloudbot project
will grow with you so as you figure out
ways that you want it to run you just
tell it and it will start to learn to do
that. All right so those are all the
basics I want to get into a little bit
more advanced stuff now. All right so
next I want to talk about model
selection because that's probably one of
the biggest open questions that people
have right now is what model should I be
using? What should be the primary model?
What should be the fallback model?
Should I use multiple models? Do I use a
local model? And the answer is yes, all
of it. And I've set up my model routing
to be pretty sophisticated. And I want
to show you. And the way I want to show
you is by asking my cloudbot. So I'm
going to say, show me your model
selection logic hierarchy. So it's going
to read from its own files and show me
right now. So the primary model I'm
using right now is Cloud Sonnet 4.5. And
that is because it is a workhorse. It is
much cheaper than Opus. And so for the
default interaction through chat, that
is what I'm using. But I also have a
fallback chain. I use Gemini 3 Flash,
which is incredibly fast and cheap. I
have Opus 4.5 here. I have Open Router
that I'm using to route to other models
that maybe aren't on my initial list.
And I was even running local models, but
I'm not doing that in this moment. And
so you can have local models running for
things like cron jobs and other really
basic tasks that don't require frontier
models, unless of course you can run one
of those frontier open source models
locally. And there are a few different
ways that you can have it select which
model it's using. You can simply say in
natural language, switch to Sonnet 4.5.
Hit enter and it's going to switch to
that model for this chat. It will also
make a decision about which model it
should use at any given time depending
on the task complexity. So we say done
back on sonnet 45. You can also type
slashmodel and if you ever get
disconnected or you run into a rate
limit on one of your models, use /model
to show or set the different models that
you have. So the way I'm doing it is for
anything coding, I'm usually using
sonnet, but if it's an especially
complex coding task, I tell it to route
to opus 4.5. For most kind of basic
interactions, I'm either using Sonnet.
Sometimes I'll switch to Haiku because
it's even faster and much cheaper. And
of course, I'm experimenting with models
all the time. And remember, the model
you select definitely affects the
personality of your Claudebot. Now, if
you don't care about that, don't even
worry about it. Just use any model
that's best for your use case. But if
you do care about the actual
personality, you're probably going to
get familiar with one model, and
switching is a little bit jarring. Now,
Clawbot packages up everything about its
personality, all of your memories, and
will send it to each of these models
individually, but the way that the
models themselves handle all of that and
respond to you still will be different
depending on the model. Okay, now I want
to get into skills and tools. This is
really what makes Claudebot special.
This is the thing that sets it apart
from any other tool that I've ever used.
It just so easily plugs into any
service, any product that you're using.
I have it plugged into my Gmail, to my
Drive, to my calendar, a sauna, Slack,
HubSpot. A majority of services that I
use, I allow Claudebot to plug into.
Now, I'm fairly technical, so I really
have thought about the security
implications about connecting all of
these different services, and I'm doing
so in a secure way. I'm going to talk
about security later in the video, so be
sure to stick around for that. But to
plug it into a new service, it is
simple. All you have to do is tell it.
That's it. It'll look for skills. If it
doesn't already have the skill, it will
write one itself. You might have to grab
an API key from somewhere, but that's
about it. You do it all through chat
with natural language. And you can even
get very sophisticated with your model
selection. So, I'm a cursor user and I
installed cursor agent in my hosting or
VPS and I'm actually able to control my
cursor agent through Claudebot through
Telegram. So, anytime that I have a
complex coding task, I'm telling my
Claudebot to delegate it out to my
cursor agent and it does that and it's
super impressive. Now, the only thing to
remember with that is that cursor agent
has basically no personality. So
sometimes when it responds back to you
and Telegram that it's done or giving
you an update, it basically strips all
the personality that you're familiar
with. And again, it's a little bit
jarring, but feel free to plug in the
tools that you're most comfortable with.
And I would recommend starting with more
lowrisk tools. Now, if you want to
browse new skills or you want to get
inspiration for things that your Cladbot
can do, go to clawhub.com. This is the
official Cloudbot skill repository by
the Cladbot creator. So up here, click
skills, and you can just see a ton of
different skills that are available.
Now, keep in mind, these are written by
other people. There's always a chance
there's some malicious code in there.
So, what I like to do is tell my
Claudebot, go download this, but first,
scan it, make sure there's nothing
malicious in there, and I want to always
use the best possible model to do that
because the best possible model is going
to be less susceptible to prompt
injection. Okay. Next, another core
mechanic of CloudBot is it being able to
do things for you on a scheduled basis.
You can have it be on a recurring basis.
You can do a one-off scheduled task. And
all of this is going to be done through
cron jobs. And if you're not familiar
with that term, it basically just means
a scheduled thing to do. And you simply
just tell it to do so. So, in 1 hour,
remind me to drink water. I'm going to
hit enter. And then it's going to create
a cron job for me. And 1 hour from now,
it's going to message me back again. and
tell me to drink water. And these can
get as complicated as you want. So, for
example, one thing that I did was I have
kind of a weird trash pickup schedule at
my house. It happens every Monday and
every other week. It's some rotating
recycling. Sometimes it's cans,
sometimes it's bottles, sometimes it's
paper. And I could never remember what
it is. So, I actually just told my
Cloudbot, "Here's the schedule." like
just took a picture of it from the
recycling company's website, uploaded
it, and say, "Look at what the schedule
is, and every Sunday, I want you to tell
me which cans to take out." And now it
does that. It tells me exactly which
recycling to put out on Sunday. Okay.
Now, I'm going to show you one of the
most powerful unlocks that I have found
with Claudebot, and that is using
Telegram groups. Now, previously what I
was doing is just DMing with my Cloudbot
directly. And that was fine, but we
would just have infinite history of
conversation. But if I wanted to have
multiple tasks running in parallel, it
was very difficult. I would say, "Okay,
hold on to that idea. Let me come back
to that and let me switch back to this
other task I had you doing." And it was
just so complicated. And then I figured
out that you can actually use Telegram
groups. And with a Telegram group, you
can set up different topics. And so you
can have a bunch of different
conversations going in parallel with
your Claudebot. And that is beneficial
in multiple ways. One, you can have a
very seamless conversation in each of
the topic channels. And that allows you
and Claudebot to stay on topic and not
get confused about other things going
on. That also saves on context window
and memory because instead of loading up
the entire conversation history which
might include days, weeks, even years in
the future of conversation history, it
loads up only what is in that topic,
saving you room in the context window.
So here's an example. I have one for
video research. I have Twitter research.
I have an ebook I'm putting together. We
have content analysis. We have a cloud
skills topic and so on. And once I find
that I'm done with a conversation or a
topic, I go ahead and just delete it.
And to do this in Telegram specifically,
you just set up a new group. You add
Claudebot as the only user. You make
Claudebot an administrator of that
group, and it should just work. And now
you also have to tell Claudebot to reply
to every message in the group chat, not
just ones that is specifically tagged in
because it's thinking it's going to be
in a literal group chat with maybe
dozens or hundreds of people. So, it's
only going to respond to messages that
it is specifically tagged in. And you
don't want that with this setup. You
want it to respond to every single
message. So, just tell it to do that and
it'll update its settings. All right.
Here's another advanced technique that I
use. As you're using CloudBot a lot, it
is storing so many memories about you.
It is creating skills and just basically
creating all of this potential clutter.
And so what I have it do daily, listen
to this. Set up a daily review of your
main files including agents.mmd memory,
anything that you can suggest. Then
propose changes here and see if I want
to make them. Basically, it is a full
audit of its codebase every single day.
So here it says, okay, I'm going to
review agents.mmd, memory.mmd, tools,
soul, identity, user, heartbeat, and
then it's going to look for things like
outdated info, conflicting rules,
undocumented new workflows, lessons from
recent failures, etc. And anytime it
finds anything that it thinks, hey, I
should actually update that or clean
that up, it's going to recommend it to
me, and then I'm going to decide if I
want to move forward with that or not. I
definitely recommend doing this just to
keep your Claudebot nice and tidy. And
you can also have full multimedia
capability. So you can have it create
videos and images and voice, anything
that you want. Here's an example. I have
it plugged into Nano Banana. And I'm
just going to say create an image with
Nano Banana of a lobster. Okay, so I had
to go back and forth with it a little
bit, but here it is. Picture of a
lobster. And again, you can give it
voice capabilities with something like
11 Labs. It can read images. Just drag
and drop in Telegram or whatever chat
app you're using. So, the capabilities
are truly endless. You just need to tell
it what you want to do. But now, here's
the really important part. Let's talk
about security. I have thought a lot
about security best practices. I am far
from a security expert. So, if you have
any thoughts about any of these, please
drop them in the comments below. So,
we're going to create a new topic in
Telegram and we're going to call this
security. So I will tell it something
like never store an API key or token
anywhere but A.Env file and never
include AENV file in your git. Now it
should do this automatically, but I like
to reinforce that. And so you're going
to want all of your API keys, all of
your tokens all in one file. And then
you don't want to include it in your git
because when you push your Git to your
GitHub account or you save it somewhere
else, you don't want that exposed in the
Git history. Now, if a lot of this
sounds very technical, don't worry. Just
type this in and that's all you need to
do. Also, OpenClaw Clawbot comes with a
security checkup feature. So, you simply
type OpenClaw security audit. So, you
can do this directly through Telegram,
but I'm going to go back to Hostinger's
terminal and we're going to do it there.
So, from the root directory, you're
going to type openclaw security audit
and then hit enter. So, here it ran a
security audit and we see two warnings.
trusted proxies missing and chmod 700.
So go ahead and fix those. And so you
can go ahead and fix those by typing
openclaw security audit- d-fix. So I'm
going to go ahead and hit that. And
there it fixes it automatically for me.
Next is just how you're hosting your
cloudbot. And that is the beauty of
hostinger. You put it on hostinger and
it is completely isolated from the rest
of your environment. It is sitting in
its own virtual private server all by
itself. It doesn't have access to other
things in your keychain on your local
computer. It can't just take over your
computer and delete a bunch of files. It
is all isolated there. And that is super
important. The next is I want to explain
the concept of clean versus dirty data.
Think about anything coming from the
internet, anything coming from outside
of your completely enclosed system being
dirty. Meaning somebody, a malicious
actor could be trying to write a prompt
injection. Here is an example. If you
have your Cloudbot reading your emails
and somebody knows that your Cloudbot is
reading your emails, they can send you
an email with a prompt injection in the
body of the email. So your Clawbot goes
to read the email, gets prompt injected,
and then reveals sensitive information.
And so just keep that in mind. I'm going
to say it again. Any time that your
Cloudbot is exposed to dirty data, it
has the potential to be malicious. So,
be very wary. And there's a few things
to do about that. One, again, just be
mindful and try to limit exposure to
dirty data. Also, the better model you
use, the less susceptible it is to
prompt injection. So, Opus 4.5 is going
to be much less susceptible to prompt
injection than Haiku would be. Also,
Claudebot has prompt injection detection
built in, but it's not perfect. It never
will be perfect because that is the
nature of AI. Another security tip, make
sure you are updating frequently. Just a
day ago, OpenClaw updated with a bunch
of new security features. Next, don't
trust skills. Claw Hub has already made
a bunch of changes to have better
security around skills that people
submit, but ultimately that is dirty
data. It is somebody submitting a skill
and there were a lot of crypto scams
already in the early days. Try not to
trust skills for as much as you can and
ultimately your Claudebot can write any
skill that it needs. It doesn't actually
have to download a skill from somebody
else. So if you can write the skill
yourself, if you can have your Cloudbot
write the skill itself, always do that.
And be thoughtful about the integrations
that you're adding. If it's something
that has highly sensitive data and you
really don't want it exposed ever, you
want zero chance of that and it's
connected to external services that
might be prompt injected. Just be very
mindful about it and you can always
limit your exposure by limiting the
integrations that you're using. But
again, the trade-off is capability
versus risk. Last, on the security
front, for any complex coding task or
really any complex task at all,
especially if there are going to be
changes to files or changes to different
integrations that I have, I always have
my Claudebot propose what it's going to
do before actually doing it. So, it's
kind of like plan mode. Think about what
you want to do. Think about the best way
to do it. Tell me what you're going to
do, and I'll let you know if I want you
to do that thing or not. All right. So
all of this is best practices, how to
get the most out of Cloudbot, but what
am I actually using Cloudbot for? And so
I wanted to show a few examples of real
use cases that I'm doing and how I set
them up. So check this out. So one, I
frequently have ideas for video topics
and I have an entire pipeline that I use
to track video ideas to research them
and I basically automated the entire
thing with Claudebot. And so here's what
I do. I drop a link in Telegram and that
link is something I want to talk about.
Whether it's like an Xlink about some
model release or a website that I wanted
to talk about that looks interesting, a
video, anything at all. Telegram throws
it to Cladbot. Cladbot does research on
that topic using the Brave API. It will
also check Twitter for trending tweets
about that topic using the Grock API.
Then it will output an ASA task for me
with all of that information. so my
entire team can see it at all times and
that ends up in the video ideas bucket
in ASA. Super useful. One drop of a link
and it's all done for me. Next, I'm
frequently checking my YouTube
statistics and I want to know how my
videos are performing and history of
videos. And so I gave my Cloudbot access
to my YouTube data API and analytics
API. So I set up a topic in Telegram
like I showed you earlier. I ask in
Telegram, okay, give me, you know, the
last three videos, how are they
performing? Cloudbot goes fetches via
the YouTube API and the analytics API
and it posts back to me in Telegram and
I can actually have it post to Slack and
tell my team how specific videos are
doing. Very useful, very easy. It's
basically like an analyst at all times
in Telegram. Next, I have it do meeting
prep for me every single day. In the
morning, I have it look at my calendar,
find all of the meetings, filter by
anybody that I'm meeting with that is
external, meaning not part of my
company, and not just some event that I
have to remind myself about. Go look up
information about that person, how we
met, what the context of the meeting is,
and then give me a summary each morning
about everybody I'm meeting. And so, the
way I do that is I have it with a crown
job in Telegram each morning. It fetches
the Google calendar. I have it check my
Gmail and I only have it check emails
from people that I know so to keep it
very safe and not expose myself to
prompt injections. And then I have it
report back to me in Telegram. All
right, so those are incredibly important
automations that I've set up with
Cloudbot. And so that's everything. I
know this was a long video. I know there
was a lot to it, but this is honestly
one of, if not the most useful pieces of
artificial intelligence that I've ever
used, even more so than Chai GPT. And so
that's how to get the most out of it.
That's how to set it up with Hostinger
and be super secure using a VPS. Thank
you again to Hostinger for sponsoring
this video. Again, go get your
HostingerVPS. hostinger.com/matthewb.
Make sure to use the code matthewb. You
get 10% off and it is a one-click
install. They've been a great partner.
I'll drop all the links down below.
Please check it out. It really helps the
channel if you go use them. If you
enjoyed this video, please consider
giving a like and subscribe.