How to USE the #1 Personal AI Assistant (Step-by-Step Setup)
551 segments
What I'm seeing a lot with founders,
they're using AI to do a lot of dumb
things really fast.
How can we use AI to do the things that
are the normal functions of the business
way better, way faster?
>> Everybody will have an AI that is an
assistant.
>> It really feels like my chat GBT gets to
know me.
>> Hermes agent is one of the most powerful
AI tools in the world right now.
Most people are using AI the completely
wrong way. They're using it for building
websites and voice agents and things
that they actually don't need. The
entire point of AI is to make you more
productive. But funnily enough, most
people use it in a way that makes them
unproductive. Just because AI can do
something doesn't mean that you should
do it. And if you're looking for the
next flashy thing to hit your dopamine,
then this video isn't for you and you
might as well click off. But if you want
to use AI to actually become productive,
then this will blow your mind. In this
video, I'll show you how you can set up
your own personal assistant that you
would usually pay $4 to $5,000 a month
for. I'll show you how you can do it in
10 minutes without needing to be
technical, and I'll skip all of the
fluff that you actually don't need.
Let's not waste any more time. Let's get
into it. Let's start with the most
important thing, which is the right
setup. Most guides on YouTube show you
complicated setups that requires you to
be technical, but this video will be the
exact opposite of that. Our personal
assistant needs a computer that's turned
on at all times, and that's what I've
set up right here. And I'll show you how
to do the same in 10 minutes. And it's
literally a one-click install. And the
entire setup will be done with our words
like chatting to chat GBT. We can do
that to host our virtual computer. And
to set it up with only chatting, we're
going to use this platform right here.
That's called abacus.ai. But we're not
going to use the regular chat. We're
going to set up our own computer. We do
that by just going top left into
supercomput. And don't worry, there's a
terminal here on the right hand side,
but we won't really be using that. Now
we have a virtual computer that lives
somewhere else, which means we can have
it so it's always on. That becomes very
crucial later when we set up the
personal assistant. When you're on this
page right here, then it's literally a
10-minute setup and it's a one-click
install. You've probably heard about
OpenClaw, maybe also Hermes agents.
Right now, OpenClaw is kind of dead and
the best personal assistant you can have
right now in my opinion is Hermes
agents. So, we can just click on this
and you can see it says set up Hermes
agents and right off the bat we're going
to have AI credits. We're going to have
web search and I'll also show you later
in this video how we can add our own
OpenAI account to have really cheap AI
credits. That is crucial when you want
to run this. Another thing that's
important is that we want to turn on
this right here, which is to have our
computer always on. We want to be able
to text our personal assistant in the
middle of the night and still have it
respond. So, I'm going to turn enable
always on. Click continue and then click
next. And now you can see it says
setting up the agent. It takes 30 to 60
seconds. And now we are done. Usually
setting up a virtual machine and
installing Hermes would take you maybe
30 minutes to an hour. We did it in one
click. But if you get this next thing
wrong, then your credits are going to
cost you 10 times as much. We want
access to the best models possible
without having to pay a ton for credits.
Let me show you the best possible setup.
Click on this little Hermes icon right
here. And what you'll be able to see is
that you can either configure it with
Claude or with the codec. What's worth
knowing, however, is that Claude
recently changed it. So you cannot use
your Claude subscription. you need to
pay for an API key which means cla is no
longer valid. Instead, we're going to
use the codec cli. So, click on that.
And now you can see it says sign into
openi codeex. It's going to give us a
URL which is this one right here. So,
we're going to copy this, place it into
our browser where we are logged into
chat gbt. Going to log in with this
account right here. If it's not enabled
on your chatbt account already, you need
to click right here. It is going to open
the settings panel. Scroll all the way
down and turn on enable device code
authorization just like this. And now go
to this link again and it should work.
Click continue. And now you can see we
need to insert the code which is this
code that we've gotten right here. So
I'm going to copy this, paste it in,
continue. And there we go. Signed into
Codeex. And if we go back, you can see
it says added OpenI Codex or
credentials. Just like that. We can then
click next. And now it's setting up the
codeex CLI. Basically the way this works
is that it's now going to use our chat
GBT plus subscription and all the
credits that is connected to that. and
you have a lot of credits inside of your
regular chatbt subscription. So if you
have one already, I'll highly recommend
that you use the codec cli. Now we can
enter full screen right here, which is
going to open up the full screen for the
MS agent. And the next thing we want to
do is of course switch the model to use
our oorthth instead. So what I'm going
to do is that I'm going to say please
use the model GPT 5.6 and change the
model so it uses our codeex
authentication CLI. Code type is going
to improve that prompt. Write it in like
this. We can now hit enter. You can now
see that the model it's using is GBT
5.5. And if I go to this section right
here where it says model, I can choose
whatever GBT model that I want to use.
Like if I wanted to do something
extensive, I could use the newest 5.6
soul model and switch to that if I
wanted to. And you can see this is using
the OpenAI codeex that we authenticated
with, which gives us a lot of credits.
This is the best way that you can set up
your personal assistant without having
to pay $200 a month for claw. Now we're
ready to build out our personal agent so
it can actually do stuff. Very excited
about this. A mistake that many
beginners make is that they create an
agent but it lives on their computer
which also means that they can only use
it when they are sitting at their
computer. But the idea of having a
personal AI assistant is that it runs
all the time and that you can send it an
idea in the middle of the night from
your phone and it'll pick it up and
actually do something. But the entire
idea of a personal AI assistant is that
you can send it a message in the middle
of the night from your phone with an
idea that you have. So we need to do a
bit more setup. But don't worry, it is
really simple. We can just say please
enable Telegram so I can contact my
agent from Telegram. I'm just going to
dictate this instead. And then we hit
enter. And because we already have a
pre-built Telegram integration inside of
the MS agent, we can now enable Telegram
very very quickly and run it all from
our phone. It's going to ask you
permission to run this command. So we
can use our arrow keys to go down and
say allow in this session and then hit
enter. And now I can see it says to
enable telegram I need your telegram bot
token from buttf father and the telegram
user ID. I'll of course show you how to
get that from your phone on telegram
search up botfather and this is the
account that's used for creating
telegram bots. Now write slash and then
new butt like this. It now says all
right a new butt. How we going to call
it? Please choose a name for your butt.
I'm going to call it wyb butt and then
hit enter. You can see we need it to be
low capital like wordyb butt like this.
That name was Ting. We can try it like
this also. Ting. Maybe just do it like
this. Albot assistant bot. Everything is
Ting. Do it in just one word. There we
go. It now says done. Congratulations on
your new bot. You'll find it. Add. And
then you have the link. And now we need
this token right here, which is token.
So only we can access it. So we're going
to copy this token. And then inside of
the Telegram, we can paste it in here
and say this is the token. And then just
hit enter. is now going to use their
secret token and set it up so only we
can contact our agent which is very
important for security. Last thing you
want to do is give it your telegram ID
and when it's done it's going to set up
Telegram for you. And the last thing you
need to do is that you just need to go
to your bot and click start like this.
Next thing you want to do is that you
want to search up user info bot and
click this one right here and then just
click start and that's going to give you
the numeric user ID. And that is all we
need to do. The agent is now going to
set all of this up for itself. So we can
very soon contact it from Telegram.
There we go. Telegram is now enabled and
verified. And now I received a
notification. And there we go. MS
Telegram gateway is connected. You can
now message this bot to reach your
agent. We can write, hey, can you hear
me? And you can see we receive a
response. Yes, I can hear you. Now we
can contact our personal assistant
anytime from Telegram. Isn't that cool?
This becomes crucial for the later parts
of this setup. This next trick will
literally save you hours because you
don't only want to be texting your voice
agent. You also want to be able to send
voice messages with long ideas that you
have and you want the agent to be able
to understand all of that. You could
either have something like a local
whisper model. However, for the most
accurate and fast results, I recommend
using something like 11 Labs. Just go to
the 11 API, click API keys, create a new
key, and I'm going to call this MS
agent. And then I'm going to unrestrict
the key and create an API key like this.
Now we can copy this API key. Go back
into our Hermes agent. Go into keys and
then I'll just search for the 11labs API
key and we can set it. Paste this in and
click save key. There we go. Now we want
to create a new chat session just by
clicking new chat. And then I'm going to
say I just added a 11 Labs API key. I
want to enable voice messages for
Telegram and I want to use the Scribe
API from 11 Labs to transcribe the
messages so you also understand those.
Can you please set that up? Thank you.
That's going to improve the prompt.
Write it in like this. And now we can
hit enter. And the thing that's so cool
about these personal AI assistance is
that I can actually just tell it to
improve itself and it does it. One thing
that's very important is that when you
add your API keys is that you do it
through the keys tab right here. You
never want to give your API key directly
to the AI. It's much more safe doing it
the way that I showed you. It asks us
for approval. So, we can say allow for
this session and hit enter. And it's
going to build out this entire
integration itself. Awesome. This is now
fixed. I can see that it just sent me a
message right here where they actually
sent me a voice message. Voice messaging
and 11 Labs Scribe transcription are now
enabled.
>> That's cool and all. However, I want to
receive text and then be able to send
the voice messages, right? So, I'll just
say, "Please enable it so I can send
voice messages." And you understand? But
you respond back with regular text.
Paste that in. This should just be an
easy setting change. And then we should
be good. There we go. This is now set
up. I can now send a voice message.
Hello, can you hear what I'm saying?
Hello, can you hear what I'm saying? It
says it back. But we wanted to be able
to respond, of course. And there we have
it. Yes, I can hear your message
clearly. Hello, can you hear what I'm
saying? It's a good idea. We could also
disable it if we don't want to. So we
can actually hear like we can see what
it wrote. But I actually kind of like it
because if for example the transcription
gets something wrong I have time to
change it. If it doesn't do this then I
don't know if it gets something wrong.
Perfect. Now Telegram is enabled as well
and I can close down this tab right here
because this is running at all times and
we have an agent that's live that I can
message from my phone at any time. But
this is where it gets really interesting
because now we have the base set up.
It's time to actually give our AI
capabilities so it can do stuff for us.
Let's get into it. Now comes the most
important part of this entire setup,
which is setting up our agent with the
things that we actually want to automate
and the things we actually wanted to do.
And this is the exact way that you
actually become productive using AI
instead of unproductive. And I'll show
you a tool, one tool that will allow you
to integrate into absolutely everything,
all the tools that you use. First thing
we're going to do is that we're going to
go inside of keys and then we're going
to scroll all the way to the bottom
where it says custom keys. I'm going to
write composio and then underscore api
and then we can click add. And here we
need to insert a value. Now go to
composio.dev. And what composio
basically is is that it's a way to give
your AI access to a bunch of different
tools. And we can get started completely
for free. When you've created the user,
then you go inside of API keys on the
left hand side. You click create API
key. And I'm going to call this MS
agent. I'm going to allow it to write
all. And then click create. And this
gives us an API key right here that we
can copy and then go back into our
Hermes agent. Paste this in and click
save. There you go. Compose your API key
is now saved. Now we go inside of chat.
I'm going to create a new session. And
then we're going to integrate into our
first tool. All I have to do is
literally just say I want to give my
agent access to Gmail as well. And I
want to do that through the Composio
API. I've added an API key that we can
use. set it up and give me a link so I
can authenticate with my Gmail and so we
can reply to all of my emails with the
Hermes agent. This will now improve the
prompt. I'm going to codeype it. It's
going to write it out and then we're
going to hit enter. While it's doing
that, let me show you a bit about what
we can do with Composio. Composio
basically allows you to integrate agents
into a bunch of apps like Gmail, Google
Calendar, Slack, GitHub. There's a bunch
of integrations that you can integrate
with. Basically, all of the tools that
you use and they have around a thousand
apps inside of here. So, by using
Composio, we only need to have one API
key and then we can give our agent
access to all of the tools that it
needs. You can see it's asking for
approval right here. We're going to say
allow for this session. And it's now
going to set everything up for us and
help us connect our Gmail via the
Composio API. There we go. Now, it has
set up Composio. So, what we need to do
is that we need to click on this link
that is it has given us right here and
click continue. We can check the link
that it is composure. It says that it
blocked the popup. So I'm just going to
click allow pop-ups and redirects. Done.
Can click the link again. Click open.
Now it opens up this link right here
where we can authenticate with our
Google account. So I can just click
continue. And then I can give it access
to everything that it needs. And click
continue again. And Composio allows it
to give it access in a secure way. It
basically manages the connection for us.
So we can close this down. And I can go
back. And now I can say I just connected
my Gmail. Can you please check if the
connection works like it should and just
dictate that in. Hit enter. And now
let's see if it has the right access. It
should have because we just logged in.
And there we go. You can see it is now
connected successfully. So now we can
create a new chat. And I'm going to
change the model to be a little faster.
And now we can say, can you please check
the latest emails that I've gotten? Hit
enter. And because we have connected it
to our Gmail, it will now be able to do
just that. You can see it's using the
Composio MCP to fetch our emails. And
now I can show the latest emails that we
just received. And now we can just
connect all of the tools that we are
using via Composio. So I can say I also
want to add my Google calendar as well
also via the Composio setup. Can we
please fix that too? Paste that in. Hit
enter. And it's now going to create the
Composio connection as well. There we
go. Just like before, it's going to give
us this link that we can open and
authenticate our Google calendar. So
just like before, I'm going to
authenticate and click continue. And
there we go. Successfully connected. And
if we go back, we can now say, I just
connected it. Can you check that the
connection is still active? Hit enter.
And it should come back. And I'll say
that it can see the connection. And
there we go. Now we have also connected
our Google calendar. And this is where
it becomes magical. We can now create a
new chat. And then we can say, every
single morning at 9:00 a.m., I want you
to check my calendar and my emails and
give me a short daily briefing of
exactly what I'm doing today. And send
that to me on Telegram. Please use the
composio connection for this. Thank you.
And then we can hit enter. And the great
thing about Hermes agents is that it can
create jobs for itself. This computer
that we on right now never turns off,
which means that it's always on. What's
going to happen now is that the Hermes
agent can give a job to itself at 9:00
a.m. every single day to finish some
task. And this becomes extremely useful
for every single thing that you want to
automate. Just by knowing how to do
this, you can connect the exact tools
that you want through Composio. You can
tell the agent that you want to do this
and this and this and it's not going to
build it out for you. I hope you start
to see the power of this. Just by
knowing this, just by using natural
language, you can automate anything that
you want. And there we have it. You can
now see it says scheduled successfully
every day at 9:00 a.m. It's going to
give us this daily calendar and email
briefing. So to test it, we can
literally just ask it, please run a test
of the daily brief for today. Hit enter.
And now hopefully we should receive a
telegram message with the daily briefing
of today. And there we go. on my phone.
We just received our daily calendar and
email briefing. I can see what things I
have to do today. I can see the unread
emails. Absolutely all of that. How cool
is that? I hope you see the power of
this because when you know how to do
these two things, one connect any tool
via Composio and two, how to set up a
crown job like we just did right here,
which is basically just a schedule task,
then you can automate anything in your
life. But what should you actually
automate? Most people when they get
access to tools like claw code and like
the AI models they start building stuff
that they otherwise wouldn't need like
they start building a new website or
they build some flashy voice agent that
can do some random thing. That is not
how you should use AI and that actually
makes AI unproductive for you instead of
productive. If you're running a business
right now and let's just take an
example, you are selling pens. You don't
want to automate things that don't help
you sell more pens. You want to take a
look at all the things that are not
selling pens and that's the things that
you want to automate. So you checking
your emails in the morning, that's
something you can automate. You checking
your calendar, something you can
automate. Your invoicing process, your
onboarding process with this setup you
have right here that is running 24/7.
You can literally just talk to your
phone and say, "Please onboard this
client. They are inside of our CRM. Send
them the contract. Thank you." And then
that entire process that would usually
take you 10 to 15 minutes opening up the
laptop doing everything done the right
way every single time. That is the power
of learning how to set this up. But now
it's your turn. Don't let this video be
another one of the videos that you watch
on YouTube but never actually take
action on. Go set up your own personal
agent and automate the things that
actually moves the needle in your
business. If you want more tips and
tricks about how you can build a
business with AI, then you can check out
our community. There's $165,000 members
inside of here right now and you can
start on a completely free trial.
Charlie just closed a $3,500 a month AI
automation client from the things he
learned inside of their community.
Inside of here, you also have all my
claw skills and training on how you
build with AI. I'll leave a link for
that right below in the description.
Thank you so much for watching. I hope
you enjoyed the video. Have a wonderful
rest of your day. The elbow out.
Ask follow-up questions or revisit key timestamps.
This video provides a step-by-step guide on creating a 24/7 personal AI assistant using Hermes agents and Abacus.ai. The instructor demonstrates how to set up the environment, integrate it with Telegram for remote control, enable voice message processing via 11 Labs, and use Composio to connect the agent to productivity tools like Gmail and Google Calendar. The video emphasizes using AI for meaningful business automation—such as managing emails and scheduling—rather than building unnecessary flashy features, ultimately teaching viewers how to automate daily tasks to increase productivity.
Videos recently processed by our community