How to Build a PREMIUM Hermes Agent Mission Control Dashboard (Step-by-Step Guide)

So most AI setups feel like a black box.

You send a prompt in, an agent gives you something back, but you don't really know what happened in between. You don't

know where the work done, which model used, which task is completed, and where the output was saved. So I built something different. I just built a

something different. I just built a Hermes setup with five specialized agents working together. I'm going to show you exactly how to build this from the scratch. And if you've never worked

the scratch. And if you've never worked with Hermes or a VPS before, don't worry. I'll walk you through everything

worry. I'll walk you through everything step by step so you can follow along even if this is your first time. Now

first we're going to set up Hemis with five agents. A top level orchestrator, a

five agents. A top level orchestrator, a researcher, a content writer, a marketer, and a developer. Now each of these agents have their own role, their own rules and its own set of

instructions. Then we will build a

instructions. Then we will build a mission control dashboard where every agent locks every action it takes. You

see exactly what they are doing, which AI model they using and how many tasks they completed on that day. We would

also connect the whole system to a Discord server where every agent gets their own dedicated channel. You talk to your researcher directly, your developer directly. Your agents are accessible and

directly. Your agents are accessible and responsive in their own lanes. They

would connect them into a full content pipeline where for instance your researcher gathers information, passes it over to your content writer who works on the content, then passes it to your marketer, all coordinated by a top level

orchestrator agent that keeps everything moving together. And as a bonus, the

moving together. And as a bonus, the dashboard has a dedicated content tab where you can read every long form documents your agents produce, plus your personalized tax tracker, Chrome job overview, live activity locks, and much

more. Now, here is what makes this

more. Now, here is what makes this different. It's a powerful machine

different. It's a powerful machine control setup that runs super light, and your VPS wouldn't even break a sweat. By

the end of this video, you would have a full operational agent workflow with complete monitoring, personalized tax tracking, direct agent communication, and a dashboard that feels like lightning. Let's build your command

lightning. Let's build your command center now. Now, the first thing that

center now. Now, the first thing that we're going to have to do is of course to get a server on which we want to run Open Claw on. And here I typically recommend that you go with a VPS because it's not connected to your home network.

A VPS is just a virtual private server.

It's just a server that you rent somewhere that you can have access to because it's connected to the internet 24/7. Now, in terms of VPS providers, I

24/7. Now, in terms of VPS providers, I typically go most of the time with Contabo or with Racknet. Now the

advantage is that the price that they give to you for the specifications of the VPS is way higher than most of the other VPS providers that I've seen out there. For hemisent or for what we're

there. For hemisent or for what we're going to set up, these two are going to be more than sufficient. If you feel you need a bit of more power, you can go with a cloud VPS 20 where you would get

60B course, 12 GB of RAM, 200 GB of SSD or 100 GB of NVME and this will way more powerful for what we want to run. So for

Contabo you're going to have to pay monthly but raknet the advantage is that you pay once and you're basically covered for one year. So for instance if you go for this VPN you're going to pay $60 a year and you're going to get a VPS

that is going to have three CPU cores 60 GB of SSD 4 GB of RAM and you would be good to go. I'm going to walk you through how you can get each of them depend on which one you want to go for and you should be fine. If you want to

get any of these VPS I'm going to put my referral link in the description section. It doesn't cost you anything

section. It doesn't cost you anything extra but it does give me some credits to be able to come up with tutorials like this to you. Now to start with contabo just come here choose the one that you want to go with and click on get started. Now in the next stage very

get started. Now in the next stage very important they do have different terms that you can run your VPS on that is a contract that you want to have or how long you want to get a VPS. And what you would see is that the higher the duration that you want to keep it for

the lower the price is going to be. So

for instance if you want to lock yourself in for one year then the VPS is going to cost you 5.6 in my case a month. If you want to go for the

month. If you want to go for the flexible option where for instance you can cancel every month then it's going to cost you $7 a month and you should be good to go. If you're pretty confident that you're going to use this VPS for a long time then this option seems to be

the best. But if you just want to test

the best. But if you just want to test the waters then don't lock yourself in for one year. You can go in for the monthly option. And then in terms of the

monthly option. And then in terms of the region my recommendation would be that you go with European Union because as you can see European Union is free immediately you choose another location it would incur an additional monthly cost. And trust me, for what we want to

cost. And trust me, for what we want to build with Hammers, latency is not really an issue. You're going to be absolutely fine going with a European Union option. It's free here because

Union option. It's free here because Contabo is an European Union is a German company and that is why you have this option. Now, in terms of the story type,

option. Now, in terms of the story type, you can either go with a 200 GB of SSD where you're going to get more space or you can also go with the NVME where you're going to get less space but a faster readr speed. The SSD option

worked for me most of the time because it just gave me more space. So, choose

whatever option that works for you best and then we can proceed. Now, in terms of the operating system here, that is where you're going to have to select the image. My recommendation would be that

image. My recommendation would be that you go with Ubuntu. So, just click click on YUbuntu here. And then here, for instance, I tested the 24.04 option and this worked as well. And then after

that, click on get started. And then

here for data protection, it's typically a good idea that you have a VPS with some daily backup. So for instance, if you pay an additional $2 a month, Contabo would always take snapshots of

your VPS so that if anything happens, you can always restore your backup. Now

imagine that you build a powerful mission control dashboard over one year and something messes up and you lose everything that you build. And that is why personally I added this backup option, but the choice is yours. And

then for additional option add-ons, leave everything the same. Very

important this part here, you're going to have to create a VPS password. And

this is the password that we're going to use to connect to the VPS server. So you

can either put in your password here or you can generate a password and after that just click on next. Now when you're done fill in your personal information, make the payment and I'm going to walk you through the next steps. If you want to go with the racknet option, I can say

that I want to go with this option here because it has six CPU cores and it's pretty powerful. Click on order now. And

pretty powerful. Click on order now. And

then here they basically give you a summary of what you are selecting for the specifications of your VPS server.

Very important. Just make sure that the operating system like here is set to Ubuntu 24.04 04 and after that you're pretty good to go. Location you can leave everything as a default and then after that you can click on continue and

then here as well you make the payment and we can go to the next steps. If you

went with the comfortable option after making the payment within a few minutes or sometimes an hour depending on your location you're going to get an email like this from Contabo. Mine is in German but the layout is going to be the same irrespective of where you are. Now

from this email we're going to need two things. We're going to need the IP

things. We're going to need the IP address, which is going to be the unique IP address of your VPS server. And we're

also going to need the username. Like I

said, mine is in German, but the layout is going to be the same. Now, with this IP address and with a username, we would be able to connect to the VPS server using the password that you set up. If

you did go with a racknet option, your email is going to look like this. Now,

from this email with Racknet, they're going to provide you the IP address of the VPS server. They're also going to provide you the username here and they're also going to provide you the password with which you connect to the VPS server. And with these three

VPS server. And with these three credentials, we will be able to connect to the VPS server. To connect to the VPS server, there are several applications out there that you can use. The good

news is that whether you're using a Windows operating system or you're using a Mac operating system, both do come with inbuilt applications that you can use to connect to the VPS server. If

you're using Windows, just click on the search bar and type in Windows PowerShell. And this is how the Windows

PowerShell. And this is how the Windows PowerShell application looks. If you're

using a Mac operating system, the input application for you is called terminal.

Now, the commands that I'm going to be using here in PowerShell are going to be the same commands that you're going to be using in PowerShell. So, these are not operating system dependent. Now, to

connect to the VPS server, just put in the command ssh space root at and then you put in the IP address of your VPS server. Remember, this is the IP address

server. Remember, this is the IP address that we received from Consabo or Raknet.

So, after pasting in this, just press enter. If this is the first time that

enter. If this is the first time that you're connecting to the VPS, you're going to be asked if you want to add this fingerprint to your host. Here,

just type in yes and then press enter.

In the next stage, we're going to have to put in the password of the VPS server. Remember, this is the password

server. Remember, this is the password that we're putting when we're setting up the VPS server with Conabo. If you're

using Racknet, this is the password that basically they give to you in your email. And for the password option,

email. And for the password option, whenever you are typing in PowerShell, you do not see what you are typing when you're typing a password. So the best option is to copy the password from somewhere. Come into terminal and then

somewhere. Come into terminal and then right click. In terminal whenever you

right click. In terminal whenever you right click it paste whatever you have in the clipboard. Although you do not see this it has pasted this into the terminal. And after that just press

terminal. And after that just press enter. And then now you can see that I'm

enter. And then now you can see that I'm connected to the Contabo VPS server. So

I'm going to clear the screen. Now

whether you're using a racknet or Contabo is going to be the same in principle just with different companies.

Now next thing that we're going to do is we're going to start with the installation of Hemis agent. And to do so, you're going to come to Hermes agents website, the official website, please. If the Hermes guys are

please. If the Hermes guys are listening, I think the page is pretty the text is pretty small, right? So

maybe you wouldn't be able to see this, so I'm going to zoom in a bit. To

install Hermes, we're going to use this oneline command here. Just click on copy and then we're going to come into our PowerShell and then just right click and then it paste the installation command.

The advantage is that this is a oneline command. So it does everything for you

command. So it does everything for you automatically, which is really a breeze just like we had with Open Claw. And

after that, just press enter. Then what

is going to happen is that Hermes is automatically going to install all the dependencies that are required for Hermes agent to run that is not installed on our VPS server. So now

let's wait for this to be done. All

right. So the dependencies installation is done. Now we're going to go into the

is done. Now we're going to go into the setup. Now here you can press up and

setup. Now here you can press up and down to select whichever option you want to go for. My recommendation as stated here would be to go with a quick setup because then the whole installation is a bit easier. So select that one when it's

bit easier. So select that one when it's green. Press enter. And then what I

green. Press enter. And then what I realized is that when it gets to this part here where we're going to choose a model, it sometimes get a bit stuck. But

if you see that, don't worry. It's just

how the setup is sometimes. But the next stage, we're going to have to choose an AI model with which we want to use for Hermes. Now, here I learned that he

Hermes. Now, here I learned that he themselves have this free AI model called news portal or something that you can use where you don't have to pay anything. But if you do have a chip paid

anything. But if you do have a chip paid account, bin pro or whatever. So I'm

personally on the $20 plan for instance.

My recommendation would be that you go with the openi codeex because they allow you to use your chachi pt paid subscription on her and this is allowed.

This is not going to get your account banned. So here I'm going to go with the

banned. So here I'm going to go with the open AI codeex. I'm going to press the space bar to select this and then what is going to happen is that you're going to get this URL where you're going to have to log in to authenticate that you

are the person using this account. Now

when you open this URL we're going to have to put in this code here at a point. I'm just going to highlight this

point. I'm just going to highlight this code and I'm going to copy this. Very

important. Whenever you want to copy in PowerShell, don't use Ctrl C because Ctrl C kills the application. Use

control shift and C. And after that, I'm going to click on this URL to open it in my browser. And then what you're going

my browser. And then what you're going to do is just log into your charg account. And then we can proceed to the

account. And then we can proceed to the next steps. After signing, you're going

next steps. After signing, you're going to have to click on continue. And then

you're just going to put in the code that we copied from the terminal. I'm

referring to this code here. Click on

continue. And then you can see that the sign in is authenticated. So when I get back to the her message, you see that it picked the sign in. It says login successful. And then here we're going to

successful. And then here we're going to have to choose a model that we're going to use for this. And then here we're going to go with a GPT 5.5 because this is the latest model as of this recording. If you see a higher model,

recording. If you see a higher model, just go with that one. And in the next stage, you're going to have to choose the terminal back end. My recommendation

would be that just stick to local. As it

says here, keep current. Press space

bar. And then here, very important, we're going to have to set up a messaging platform that we're going to be using to chat with our AI agents without having to come to the terminal every time. Now, here already selected.

every time. Now, here already selected.

I'm going to press on the space bar. And

then here, it's going to list a whole lot of messaging platforms that we can use. What I like with Hermes is that

use. What I like with Hermes is that they don't give you a thousand messaging platforms that you don't need. They

really concentrate on the ones that you need. So, here we're going to select in

need. So, here we're going to select in this example in our setup, we're going to be setting up Telegram, and we're also going to be setting up Discord. But

first, I'm going to be setting up Telegram. And then when we get to the

Telegram. And then when we get to the discord pad, I'm also going to show you how we can add a discord pad. Very very

important at this step. Please do not press the enter yet. You have to press spacebar to check this that you want to set up telegram. And after that, press enter because if you press enter without

the space bar or tick telegram, it assumes that you didn't select anything.

Now here, very important, we're going to have to create a telegram bot token. And

to do so, just come to Telegram and look for a bot that is called BotFather. So

you can see that bot father has over 9 million monthly users. So you're pretty sure that you're working with the right one. Now we're going to have to create a

one. Now we're going to have to create a new bot for our Hemis agent. And then to do that, just type new and then select new bot. And then we're going to have to

new bot. And then we're going to have to give our bot a name. In my case, I'm going to call this computer mechanic Hemis, right? And then I'm going to

Hemis, right? And then I'm going to press enter. And then we're going to

press enter. And then we're going to have to give the bot a username. But

very important, the username always has to end with the word bot. So I'm just going to paste in the same name that I give to it. And I'm just going to add the word bot and press enter. Now what

you're going to see is that they're going to give you an HTTP API key and this is what is called a bot token. When

you click on this, it copies this URL or the token into your clipboard. And then

we're going to come back to PowerShell and then just right click into PowerShell. When you put in the Telegram

PowerShell. When you put in the Telegram board token, you do not see this because it's a token. It's a secret key, but this has been pasted into the terminal.

And after that, just press enter. And

then very important, what I like about Hermes is that during the Telegram setup, it lets you add yourself as the only person that can talk to your bot.

And for you to do that, you're going to have to tell Hemes agent your Telegram ID. And to get your Telegram ID, they

ID. And to get your Telegram ID, they have given you this procedure here.

We're going to message a bot that is called user infobot on Telegram. And

then whenever you message this bot, it replies with your Telegram ID. Now, I'm

going to show you how this is. Just go

to Telegram. And then in here, just search for a bot that is called And this is how the bot looks, right? And then

whenever you send any message to the bots, for instance, hi. The bot just responds and gives you your ID. And then

here you're going to copy the ID here.

We're going to come back into the Hemis setup. And then we're going to paste in

setup. And then we're going to paste in the ID. And after that, just press

the ID. And after that, just press enter. And then he is going to ask you,

enter. And then he is going to ask you, hey, do you want to set this user ID as a home channel? Just type in Y for yes and press enter. In the next stage, Hermans is going to ask us if you want to allow the system service for Hemis to

run in the background. I would highly recommend that you do this because then it's always going to run in the background without you having to start this automatically. So I'm going to

this automatically. So I'm going to press Y for yes and I'm going to press enter. And the next stage as well is

enter. And the next stage as well is going to ask you how do you want the gateway to run in the background. Now we

are not using a laptop. We are using a VPS. So press down to come to the

VPS. So press down to come to the section here. Press space to tick on

section here. Press space to tick on that. And then it's going to select that

that. And then it's going to select that you want to run this on pseudo. And then

it's going to ask us again which user do you want to run the system gateway service on here. Type root because I'm using it on the VPS. So the only user that I currently have is root. And then

press enter. And then it's going to ask you, hey, do you want to start a service now? I'm going to say yes by pressing Y.

now? I'm going to say yes by pressing Y.

And then I'm going to press enter. And

then it's going to ask you again, do you want to install the gateway as a background service here as well? Press Y

for yes. Then press enter. And then

basically the hemis installation is complete. Now what we're going to do now

complete. Now what we're going to do now is we're going to go to our Telegram board and see if everything that we did from the configuration perspective is working. But before we go to Telegram,

working. But before we go to Telegram, you can actually check to see if things are working correctly by just typing hemisphere here and then pressing enter.

And what this is going to open is that it's going to open sort of a chat in a terminal user interface. And that is going to be the first time that you're actually going to see if the configuration and the installation was set up correctly. So here I'm just going

to type hi and if the bot responds then it means the installation was successful and the AI model that we connected was also working. So here you see that the

also working. So here you see that the bot responded hey hi what can I help you with today? So it means that this is

with today? So it means that this is working. So we can just leave the space

working. So we can just leave the space by typing flash and then type exit and then press enter. And this lets you leave this terminal user interface for the chat. So what we're going to do now

the chat. So what we're going to do now is we're going to go to botfather and then we're going to start chatting with our bot by clicking on the URL that we have here and then just click on start.

Now you're going to see this weird message on command start. That's fine

because the bot doesn't really know this. But what we're going to do is

this. But what we're going to do is we're going to start chatting with the bot by just saying hi. And if the bot starts typing then this is a good sign that the configuration was done correctly. So you see that the bot

correctly. So you see that the bot responded. Hi what can I help you with

responded. Hi what can I help you with today. So our telegram bot is set up.

today. So our telegram bot is set up.

Now we can start with the main gig. To

make your life easier I have prepared all the prompts that we're going to use so that you can just copy paste and you'll be good to go. When you start chatting with the bot the first you're going to have to do is to introduce

yourself to the bot. Tell the B who you are and give the B also a name and then give the board an idea of what do you want to do with this whole hemis setup that you're doing. So what I'm telling the board is hey your name is

orchestrator. You are going to be the

orchestrator. You are going to be the systemwide coordinator for my multi- aent setup. I don't want to read through

aent setup. I don't want to read through everything but then I'm also telling the very important my name is Larry and I am the owner. I am the highest authority

the owner. I am the highest authority for whatever we're going to build. Then

I'm also going to tell the board that his role is going to be to oversee a full agent system where we're going to work with other agents or four other agents who are called scout scripe reach

dev. And then I also tell the agents a

dev. And then I also tell the agents a quick summary of what the roles of these agents are going to be. And then I'm going to tell him, hey, your job is to be the top level operational coordinator keeping the entire system stable,

scalable, and easy to manage. Now,

please feel free to adapt this prompt to exactly what you want to do with a Herman setup. And I'm going to paste in

Herman setup. And I'm going to paste in this prompt. And I'm just going to send

this prompt. And I'm just going to send this. Now what happens when you send

this. Now what happens when you send this prompt is that the agent basically updates its memory that hey the user is called Larry. And it also updates the

called Larry. And it also updates the other stuffs. And it says I'm going to

other stuffs. And it says I'm going to operate as orchestrator your top level system coordinator and all that. And it

also repeated that hey we're going to have a setup with these agents who are going to help me with my work. Now in

the next stage I'm going to send the agent my personal operating rules. Now

the reason why I do this sometimes when the agents are working they just get stuck and you do not know if they are working or not. So I like them to break whatever they are doing into multiple steps so that I can really follow where

they are in exactly what they doing and if there's something complex that they are doing I also want them to give me sort of an approval for me to approve before they proceed. You don't need to do this but I do this because I find it

very very helpful. All right so you see that orchestrator added this to his memory that these are Larry's permanent operation rules. Now in the next stage

operation rules. Now in the next stage we're going to give orchestrator sort of a heads up. We're going to tell him hey we're going to create four additional independent persistent agents. Now

persistent is very important here. And

then we're going to tell him hey the agents are going to be called scout who is going to be handle research scribe who is going to be responsible content reach dev and so on. Then I'm telling these are not going to be temporary

agents but permanent one. The key aspect here is that I want every agent to have their own soul their own identity their own memory and so on. Now when you do it like that, every agent has their own

memory that they working with. They have

their own identity and stuff like that and it really doesn't crosscontaminate the system because you don't want to be talking to dev and dev is reading the memory of scribe. It just doesn't make sense and that is why we need this

separation. So I'm going to send this to

separation. So I'm going to send this to orchestrator and he's going to build this as well. So I'm telling him, hey confirm that you understand the plan and the roles of the four agents that we're going to be creating. And then he says

confirm Larry. Understand the plan.

confirm Larry. Understand the plan.

Orchestrator. This is basically him.

Then we have scouts. We have scribe, we have reach, we have dep and then each agent will be permanent. Basically

repeating what we said so that we are sure that he understood exactly what we wanted to do. And in the next I'm going to tell him hey for the four agents that we have these are the descriptions that I want you to give them. So for instance

for the system prompt for scout I'm saying hey you are scout you're a deep reset specialist for my agent OS. I'm

telling exactly what the job or the role of scout is and I'm doing it for all the agents. Now, these are going to be

agents. Now, these are going to be written into the system files for these agents so that they know exactly what they are and we're going to test this.

When I send the agents afterwards, who are you? You're going to see that

are you? You're going to see that they're going to respond exactly with what they do. I'm going to copy this.

I'm going to send this to orchestrator and then he's going to build this. And

you can see because I told him, hey, I want to approve everything that you do.

He's saying that he gave me the plan.

He's going to inspect this. He's going

to create the persistent HMS profiles.

He's going to create the isolated workspace. And he's going to configure

workspace. And he's going to configure each of them. Do I approve this plan?

Say approved. And this aspect of the setup sometimes takes long. So when your agent is working, just leave him to work. And you can see as I gave him the

work. And you can see as I gave him the permanent rules, I can see currently where he is. So he's currently in step two of six. He's now in step three of six. And this really helps you to be

six. And this really helps you to be able to follow where they are in whatever attacks that they are executing rather than me just sitting here going dark. So he's done. You can see that he

dark. So he's done. You can see that he created this persistent hemis profiles of the agents. Each agent has their own workspace like we said and then it verifies that they all have their sole

identity user agents and stuff like that. And then here he's asking me

that. And then here he's asking me already this what I love about Hermes.

Do you already want to configure the discord? But I'm going to say not at the

discord? But I'm going to say not at the moment because there's one more important things that I want to tell the agents. I'm going to copy this prompt

agents. I'm going to copy this prompt here. I'm going to tell him hey I'm

here. I'm going to tell him hey I'm going to define how I want them to work.

If his dedicated memory he stores contest relevance to his role. So I

define what every agent stores in their memory. And one very very important

memory. And one very very important aspect is here the role boundaries. Now

I don't want dev to be writing a research report for me. So for instance if I tell them to write a research report he's going to say hey I'm not the best guy for this. Why don't you talk to scout who is basically responsible for

this. And then I also give him that I

this. And then I also give him that I want the agents to be able to remember previous conversations and build on them. So I'm going to copy this prompt

them. So I'm going to copy this prompt here and I'm going to give it to orchestrator as well. I'm going to approve this. And then one of the

approve this. And then one of the beautiful things about him is you can see that immediately he did this. He

already created a self-improvement review with a skill that we didn't even ask for and this is really one of the best parts of her. Now in the next stage we're going to create what is called a router cheat sheet. Now the idea is that

whenever I'm chatting with the agent I want to be able to use natural language and then for instance orchestrator can identify what I want and then basically dispatch it to the respective agents who is responsible for that. And also I'm

also going to let him enable some slash commands on telegram. So for instance, if I type a scout and I type after that a task, then telegram immediately knows, hey, this tax is for scout. If I do the same for dev, then he knows that this is

for dev. So I'm going to copy this

for dev. So I'm going to copy this command. I'm going to put it here and

command. I'm going to put it here and I'm going to send this as well. Going to

approve this. Now here, when he wants to run a command, he's going to ask you if you want to approve this. So here I'm just going to say always approve. If you

want, you can also just approve this once, but I prefer to go with always so that it runs automatically. Basically,

what that command was is that he wanted to restart the gateway. And because he's restarting the gateway, you see that it writes gateway restarting, your carrying tax will be interrupted. And once the gateway restart is complete, he is going

to basically pick up and continue with what he was doing. So you see that the gateway restart is complete and then he basically picked up from where he was before the restart. All right. So Krita

is technically done. He has set up the router shortcut. You can see that you

router shortcut. You can see that you can just type. So scout is going to do this. Cry is going to do this marketing.

this. Cry is going to do this marketing.

So for instance, if you type forward/cout find five recent sources on AI agents and it basically knows that this message is exactly for scout and also added some fallback. So this is

working very well. Now what I would highly recommend in the next stage is we're going to have sort of a shared team awareness. Now imagine that we have

team awareness. Now imagine that we have five people working in an office. So

typically when you're working in an office you know what the roles of your other team members are. And with this whenever you need a specific help you know which agents can help you with this task. So I'm going to make all the

task. So I'm going to make all the agents aware of the teams that they are working with so that they know what kind of setup and what kind of roles that we have in our current setup is. So I'm

going to copy the shared team awareness prompt and I'm going to send this to orchestrator to implement this. I'm

going to approve this. So confirmed

every agent has now a shared team awareness. So all the agents basically

awareness. So all the agents basically know the other agents that they working with. Now in the next stage we're going

with. Now in the next stage we're going to set up what is called a full content pipeline. Now this is optional but it's

pipeline. Now this is optional but it's extremely helpful. So for instance for

extremely helpful. So for instance for my YouTube channel if I want a topic to be worked on for instance I can start that scout researches the topic first and then scout passes that information to scribe and then scribe writes that

blog content for me and after that scribe passes it to reach and then reach creates the social media post for that particular content and after that reach also builds the marketing and promoting

promotion strategy for this and then reach would basically deliver the final plan. So what this means is that you

plan. So what this means is that you have a task and this is basically routed step by step through all the agents and then you finally get your output in the way that you want it. Like I said this is optional but I find this very helpful

for instance when I'm researching a topic for my YouTube and I basically want the A to Z of that topic. I'm going

to copy this and I'm going to send this to orchestrator to implement this. All

right. So Orchestrator is done. He's

basically set up the full content pipeline. Now let me test this out by

pipeline. Now let me test this out by just saying run full pipeline on AI automation ideas for solo creators and small teams. And let me just show you an example of what the output would look

like where all this will be run across all the agents and then we get the final output of this research. And basically

you see that it says step one starting scout to do the research and he's basically waiting on scout. So scout is done with his research and now this has been passed to scribe who is working on

writing the content. All right. So you

can see he's done with a full pipeline.

So it basically gives me the output and the research that all the agents did and reach is also asking me hey completed promotion plan. It details out what

promotion plan. It details out what exactly it wants to do. And he's saying if you approve I'm going to build a lead magnets first and basically highlights exactly what we want to do. And this is very nice because it passes exactly what

you want to do through all the agents and it brings you the final output. I'm

not going to go beyond this because that is going to get a tutorial very lengthy, but it basically gives you an idea of what the full content pipeline should look like. In the next stage, we're

look like. In the next stage, we're going to go to an integral part of the setup where we're going to work on the Discord integration. Now, this is not so

Discord integration. Now, this is not so complex, but you need to pay a bit of attention because otherwise if you miss one step, you would miss the connection.

Now, the first thing that we're going to have to do is to let me make this bigger. Just come to

bigger. Just come to discord.com/developers/lications.

discord.com/developers/lications.

We're going to have to create a discord bot for this Hemis agent. Now, when you come to this page, just click on new application and then you can name your bot whatever name you want to name. I'm

going to call this computer mechanic Hemis bot and then accept the conditions and then create this bot and we should be good to go. Now, on this page, very very important, come to the bot section

here on the left and then when you scroll down, reset the token. Now a

token is going to be generated with which we're going to connect our discord setup to the hemis agent setup. Now

click on reset token say yes I want to do this and then you're going to have to authenticate your account and then once you're done you see that the bot token has been generated for us. Now copy this token and keep it in a safe space. So in

my case I'm going to put it in notepad for the time being. Going to call this discord b token. I'm going to save this.

I'm going to paste this in here. Now

we're going to need this token when we are connecting Hemsus to Discord later.

Now once that is done scroll down here and then make sure that you enable presence intent, server members intent and message content intent. I'm going to highlight all these and then save

changes and after that just scroll down as well and then for both permissions choose administrator and you should be good to go. Now, after you have done this, come to the left section here and

go to O. And then here, what we're going to do is we're going to add this bot to our Discord server. Now, before we do that, we're going to have to create a new Discord server for our agents. Open

Discord on your PC. Scroll down and then come to the section here where you have the plus and click on add a server. And

then come to create my own for me and my friends. And I'm going to call this

friends. And I'm going to call this computer mechanic Hemis OS. And I'm

going to click on create. And then when this is done, we're basically done with the first step. Now what we're going to do is we're going to go back to the Discord developer tools. And then when you are here under o, scroll down and

choose bots. And then here select for

choose bots. And then here select for bots permissions administrator. Then

when you scroll down here, you see that the URL is generated for you. Now this

is the URL which way we're going to invite this bot that we just created into our discord server. I'm going to open a new tab and I'm going to paste in the URL here and then press enter. I'm

going to allow this because it wants to open a pop-up. I'm going to click on continue to Discord. And then what is going to happen is it's going to ask you, hey, where do you want to add this computer mechanic but that you just created? I'm going to click here. I'm

created? I'm going to click here. I'm

going to scroll down and I'm going to select the Discord server that I just created. In my case, I call this

created. In my case, I call this computer mechanic Hemis OS. I'm going to click on continue. It's going to tell me this is an administrator. I'm going to say yes. And I'm going to click on

say yes. And I'm going to click on authorite. And then this has been

authorite. And then this has been successfully created. So when you go

successfully created. So when you go back to your Discord server, they have going to do the same. It's just telling me the same. I'm going to click on continue again. Authorize. Go to

continue again. Authorize. Go to

computer mechanical. You see that you have a message that good to see you.

This board has now been added to the server. Now, what I'm going to do here

server. Now, what I'm going to do here is I'm going to clean the server a little bit. I'm going to delete all

little bit. I'm going to delete all these channels that I have here because we're going to create new channels for our setup. Going to delete this as well.

our setup. Going to delete this as well.

And then we're good to go. Now, what

we're going to do is we're going to need a few things from here when we are setting up the Discord for Hemis. First,

we're going to need the server ID of this particular server that we created.

And we're also going to need your personal user ID. Now for the server ID, you're going to click the server name here. And then you see that there is

here. And then you see that there is copy server ID here. If you do not see this, it means that developer options is not enabled on your discuss setup. And

to enable this, just come to where you have your name. Click on the user settings and then scroll down. You would

see developer options. Make sure that developer mode is enabled because if this is disabled, you wouldn't see the option to copy the server ID. Now when

this is enabled, just come to the server name. Click on the name here and then

name. Click on the name here and then you'll see copy server ID. So I'm going to paste this also in my notes. Server

ID. And the next thing that we're going to need is your user ID. To get your user ID, just click on your name here.

And then you can also see copy user ID.

If you don't see the user ID as well, it means that you haven't enabled developer options. So I'm going to save my user ID

options. So I'm going to save my user ID also here. I'm going to paste it here

also here. I'm going to paste it here because I don't want to expose this. And

then basically with the discord port token with a server ID and user ID, we would be able to connect our hem setup to this code. So what we're going to do now is I'm going to go back into PowerShell and then we're going to

connect this code to this H setup.

Before I did, I'm going to have to connect to the server again. I'm going

to SSH again. Going to put in my password, connect it to the server. I'm

going to clear the screen by just typing clear to make it a bit cleaner. And then

now to connect this code to Hemis, just put in the command Heis setup. We're

going to go through the same setup again, but now we're just going to focus on this code. And then here, we're not going to change anything. The model is going to be the same. Just press enter here. And then it's going to ask you,

here. And then it's going to ask you, hey, for OpenAI codeex, do you want to do anything? I'm going to say use the

do anything? I'm going to say use the existing credentials because that is not what we want to do. We just want to work on this code. So, I'm going to say don't change anything here by pressing one.

And then the model, I'm going to skip this. I'm going to say keep what you

this. I'm going to say keep what you have currently. And then here as well,

have currently. And then here as well, I'm going to say keep current. Here as

well, I'm going to say keep current. I'm

going to press just press enter here to use the default because this is not what we are looking for. Here I'm going to say keep current settings and then this is the part that we want to come to. Now

we can see that telegram is already configured for discord. It says not configured. So press down on your arrow

configured. So press down on your arrow key and press the space bar to check this code. And then once this is checked

this code. And then once this is checked in green press enter. And then it's going to tell you hey you have this code already configured. You already

already configured. You already reconfigure it. I'm going to press N for

reconfigure it. I'm going to press N for no because we don't want to touch Telegram. And then here it's going to

Telegram. And then here it's going to ask you for your discord boat token. I'm

going to copy my discord boat token.

Come into the partial and then right click. Like I said, whenever you right

click. Like I said, whenever you right click in partial, it basically paste whatever you have in the clipboard. But

you do not see this. After that, just press enter. And then it's going to tell

press enter. And then it's going to tell you, hey, do you want to allow specific people access to this bot and I'm highly going to recommend you to do so because if you do so, only your user account can be talking to this bot. And then here,

what you're going to do is you're going to put in the user ID that we created.

I'm going to paste the user ID here again to show you what I'm talking about. The user ID is what when you

about. The user ID is what when you click here, you see copy user ID and that is your user ID because only this user will be allowed to talk to your Discord. After that, press enter. Now,

Discord. After that, press enter. Now,

Hermes typically recommends that you have what is called a home channel where you're going to get crossplatform notifications, chrome jobs, and stuff like that. So, very quickly, let's

like that. So, very quickly, let's create this home channel. I'm going to go back to my Discord and I'm going to create a new channel and I'm going to call this Hermes home. Then I'm going to click on create channel and then we're

going to have to give Hermes the ID of this channel. Right click on the channel

this channel. Right click on the channel and click copy channel ID. And this is going to be sort of our home channel.

I'm going to go back into the setup and I'm going to paste this in there. Now

after that just press enter and then it's going to tell you hey do you want to restart the gateway to pick the changes? I'm going to say yes by

changes? I'm going to say yes by pressing Y and then press enter. and the

gateway is going to be restarted so that the new Discord configuration that we set up will be picked up. And then here it's going to give me some options. I'm

just going to say done because we don't want to change anything. We just wanted to work on the Discord. So this is basically done. So from the back end the

basically done. So from the back end the Discord is set up. Now we could have technically just given orchestrator here the B token, my user ID and stuff like that for him to do the setup for me, but it's typically not recommended to put

tokens in chat because if someone hacks into your Telegram, they basically get all the credentials. And that is why I prefer and it's always advisable to do the setup yourself and then basically tell your agent about it. So here I'm

going to tell orchestrator hey I've configured the discord again for our Hermes installation. Now update the her

Hermes installation. Now update the her setup so it can work with this particular server. So here I'm going to

particular server. So here I'm going to copy this prompt. I'm going to come here and I'm going to very important replace the server ID where I say paste your discord server ID with your server ID.

Click on your name. You see copy server ID and that is basically your server ID.

And then once you do this, what orchestrator is going to do is orchestrator is going to try to connect your Hemis setup to this particular Discord server because on this server we're going to be creating all the channels for the agents. We're going to

do all the configurations and stuff like that. So it makes sense that her knows

that. So it makes sense that her knows that hey this server exist and one of the good things that I like about Hermes is that it periodically com conversation because as you are talking to the agent there's something called a context

window which is the sort of the conversation history that you are you currently have in that chat session. as

your conversation history blows up with every message that you send, it always packs the entire conversation to the AI agent. So what Hermes does periodically,

agent. So what Hermes does periodically, which is very good, Open Claw does this as well, but I see that Hermes does this more periodically. Whenever it sees that

more periodically. Whenever it sees that the context window is bloating up, it basically compacts it so that the context window gets smaller and then it also saves you a lot on AI tokens. So

here he's basically did a compaction and he's continuing with the instruction that I gave him. Strator confirmed, he verified and updated. What I also saw is that he also sent a test message in the channel to make sure that he is connected to the server. What we're

going to do now is we're going to start creating the channels for all the agents that we have. And for that, I'm going to tell him, hey, I've made you an admin on the Discord server you just created.

Create a test channel called Hemis Test.

And he should also provide me with a channel name ID so that I can confirm that he has full access to the server.

So, I'm going to open Discord and see if he creates that server. All right, this went very well. He created a channel called Hermes Test. Now, in the next stage, I'm going to let him create all the agents channels that we need. So,

we're going to need technically four channels. One for scout, one for brief,

channels. One for scout, one for brief, one for Skype, Reach, Dev, and so on.

So, I'm going to copy this prompt. And

the next stage, I'm going to tell him, hey, delete the test channel that you created and create four new channels for the agents. So, one is going to be

the agents. So, one is going to be called Scout Briefs, Scribe Scripts, Reach Marketing, Dev Build. And for all the channels, he's also going to add some kind of an emoji prefix for the

channel just to look nicer. I'm going to copy this. I'm going to send this to

copy this. I'm going to send this to orchestrator so that he implements this.

In the time being, I'm going to open this code and we should see that the test channel is going to be deleted and he's going to create the new channels that we just told him to do. Perfect.

So, you can see that he deleted the test channels and then now we see that he created four new channels for us, one for every agent. So, this looks very good. Now, in the next stage, we're

good. Now, in the next stage, we're going to tell him, hey, for the channels that you just created, I want you to bind each agent to their specific channel. So, Scout should be bound to

channel. So, Scout should be bound to his channel. subscribe to his channel,

his channel. subscribe to his channel, read to his channel, de to his channel, and after that, I'm going to send him this. And then what we're going to do is

this. And then what we're going to do is that once he's done with this channel bindings or the Discord channel bindings, we're going to go into every Discord channel that we have, and we're going to send a search message, who are

you? If the binding was correctly, when

you? If the binding was correctly, when we send who are you in scout channel, he should respond, hey, I'm scout. And the

same thing would apply for all the other agents. Perfect. So, orchestrator is

agents. Perfect. So, orchestrator is done. He says every channel has their

done. He says every channel has their own agents binding. So what we're going to do is we're going to go to our Discord server and we're going to start with scout. We're going to send a

with scout. We're going to send a message who are you. And if the binding was done correctly, we expect him to say I'm scout. Perfect. So he says I'm scout

I'm scout. Perfect. So he says I'm scout Larry's research and he also mentions the other teams that he's working with.

So I'm going to send this to all the agents on all the channels just to be sure that the all the agent specific bindings are correctly done. Now this is a very important step because we want to be talking to the agents in their

dedicated channel and that is why we have to make sure that every agent is behind the right channel. So perfect

scribe responded in his channel reach responded in his channel and dev also responded in his channel. So the code integration works the channel to agents binding also worked perfectly. Now the

next thing we're going to slowly start to build the mission control dashboard.

Now it's very important that we can monitor what the agents are doing. So,

we're going to build a local agent login system. The idea is that whenever an

system. The idea is that whenever an agent finishes a task, we want them to log their name, the task that they did, the model that they used, the status, whether it was completed or failed and when they performed that task. And by so

doing, we would be able to have a very good overview of what the agents are doing every time. So that if something fails, we can know exactly what fails or if we're agent supposed to perform a task and we see that the agent is idle,

we know that the agent didn't work. So

we're going to create this sort of local agent login system and we're going to build this. This is going to be hosted

build this. This is going to be hosted on a VPS server but it's going to be an extremely light server. We're going to do this using SQL light. So it wouldn't take any tool on your database. All

right. So orchestrator is done. He built

the database. He built also the scripts that the agents are going to be using to do the login. And also here just an example of a test smoke test that he did. You have the agent's name. You have

did. You have the agent's name. You have

the exact thing that they built. We also

have the status of the task. And we have the time stamp. The only thing that I see here is that the model writes unknown. So I'm going to tell him this

unknown. So I'm going to tell him this looks good. But I think the model used

looks good. But I think the model used is not working. It it writes saying can you check why and fix this. So this is now fixed. So you see that now we have

now fixed. So you see that now we have the the agents name the task they performed the model they used and so on.

Now what we're going to do is that we have now built this agent login system but the agents have no idea this actually exists. So what we're going to

actually exists. So what we're going to do is we're going to roll out this login activity to all the agents so that they know that this success. So I'm going to copy this and I'm going to send this to orchestrator to implement this. All

right. So the agents now know that we have this agent login activity system.

Now very very important because we have a database set up on our VPS where everything the agents do they are going to log. We need to have a maintenance

to log. We need to have a maintenance system where we're going to clear or flush out old login activity because we don't need activities from the agent for more than 3 months or something like that because this will just bloat up

your VPS. So, we're going to set up sort

your VPS. So, we're going to set up sort of an automated cleanup to make sure that things that are older than 30 days for instance are cleared from our database because we do not need this. If

you want, you can also make this that things that are older than 7 days can basically be deleted so that the database stays clean and your VPS stays very light. So, I'm going to send this.

very light. So, I'm going to send this.

Perfect. So this is done as well. Now

we're going to move to the important aspect where we're going to build the mission control dashboard. And for

mission control dashboard because it requires coding, we set up a dedicated agent dev who is basically going to be responsible for coding. Now we're going to start by talking to dev. Let me just prep tips make sure that he's ready.

Hey, are you ready for some big task?

Now what we're going to tell dev is that I want to build a readonly mission control dashboard for our hemis setup.

And I'm going to tell him that the dashboard is supposed to be stored here.

And this is the port that I would like to use to access the dashboard through this IP only. And then also very important, we're going to tell him that the dashboard is going to serve some roles. We're going to be pulling the

roles. We're going to be pulling the agent locks from the agent database that we just built. We're going to also be pulling some sessions on token user and stuff like that. And this is already built in by her. We're also going to be

pulling a caban database which is also autocreated by Hemis for us to basically see what kind of Chrome jobs are there and stuff like that. And also we are going to be pulling in some other aspects to pull some database from the

gateway status and all this is already ready in hemis. So I'm going to send this to dev that this is what we intend to build. All right. So dev is done with

to build. All right. So dev is done with that. He basically was able to locate

that. He basically was able to locate all the end points that we would be interested in for the machine control dashboard. And this looks very good. Now

dashboard. And this looks very good. Now

what we're going to do next is we're going to let him set up the full backend for the mission control dashboard. The

mission control dashboard we're going to load this on a Python server. And we're

not going to be using any complicated things that would slow down the mission control dashboard. It's going to be a

control dashboard. It's going to be a simple Python server that is going to be served on the spot and it's going to serve just one file index.html which is going to be the mission control web page. And then we're also going to

web page. And then we're also going to be pulling five data sources. Like I

said, we're going to be pulling the gateway data. We're going to be pulling

gateway data. We're going to be pulling the activity data from the agents sessions, VPS help, Chrome jobs, and stuff like that. So, I'm going to copy this prompt and I'm going to send this to dev. If the prompt is too long and

to dev. If the prompt is too long and you do not have a paid Discord subscription, they're going to tell you that the prompt is too long. You can

only send this as a file. But this is absolutely fine. If you send it as a

absolutely fine. If you send it as a file, the agent will still be able to read this. So, you do not need to pay €9

read this. So, you do not need to pay €9 or $9 a month just to be able to send long messages to the agents. All right,

so dev is complete. He built and started the mission control back end. So, you

can see that the Python server is saved.

We have the index.html file saved. All

the pots, all the things that we need are also pulled in from the back end.

And he's saying that the server is running at this spot. Now, what we're going to do is we're now going to start to build the skeleton of the dashboard.

So, I'm going to tell Dev and I'm going to be explaining what we're going to be doing. So, I'm telling him, hey, build

doing. So, I'm telling him, hey, build the index.html skeleton from the dashboard. And I'm giving dev a visual

dashboard. And I'm giving dev a visual orientation of how I want the dashboard to look. Now, this is my design. You do

to look. Now, this is my design. You do

not need to use this exactly. If you

have any design inspiration that you want to go for, you can also replace this with the prompts that I'm giving him. And I'm also telling him, hey, for

him. And I'm also telling him, hey, for the skeleton that you're going to be building, we're going to have some tabs like overview, agent, tax, schedule, content, and all of that. Now, this step

is very, very important because when you designing a page or a web page, you want to make sure that the fundamental designs are correct, you have the right font, you have the right background, you have the right layout before you start

to populate the dashboard. Because if

you start populating the dashboard and you don't like the design and you have to rebuild everything and it's really a mess and that is why I like to set the foundation where I have a skeleton, I have the frame of how I want the design to look and then from that you can

basically build on that and it becomes a very easy way to build the dashboard. So

def is done. The beautiful thing is that you see that he even added a screenshot that I can see how the dashboard looks and this is one of the things that I like about him is that it's not currently there with open claw. So now

I'm going to tell him how do I access the dashboard? How do I assess the

the dashboard? How do I assess the dashboard? Now, we're going to do this

dashboard? Now, we're going to do this through what is called SSA tunneling, but I just want to do this so that you can see in case you do not know how to access it, then dev is also going to show you how we're going to do this. So,

like it says here, basically this is the part that is going to save the dashboard. Now, to access this, we're

dashboard. Now, to access this, we're going to use what is called SSH tunneling. So, just copy this command

tunneling. So, just copy this command that we have here and then open Windows PowerShell. Now once powershell is

PowerShell. Now once powershell is opened here because I already have this I'm going to click on plus and then just put in this command ssh space minus l and then this is going to be the port on which the mission control dashboard has

been saved and then here this is going to be your username and then just replace the section where the IP address of your VPS and after that just press enter. Now you're going to be prompted

enter. Now you're going to be prompted to put in the password of your VPS server. So I'm going to copy the

server. So I'm going to copy the password. I'm going to paste this in

password. I'm going to paste this in here and I'm going to press enter. Now

it's going to look like this. It's going

to log you in again into the VPS button, the SSA tunneling is already active. So

when I go back to my browser and I open the address that dev give to us, it should basically open the mission control dashboard. Beautiful. So now you

control dashboard. Beautiful. So now you see that the dashboard has been loaded.

This is basically our mission control dashboard. Now we're going to continue.

dashboard. Now we're going to continue.

The design looks fine. What I personally would prefer from a design perspective, I like the sort of dotted kind of background because it's just the vibe I want to go for. So, what I'm going to do is I'm going to send this to Dev and I'm

going to say, "Hey, can you make the background look like this?" Nice. Now,

you do not need to do this, but like I said, this is the kind of vibe that I want to go for. And this is why it's very important that we set the foundation. Because here, if the font

foundation. Because here, if the font sizes are too small, you can tell them to increase the font sizes. If the

background is not what you want to go for, you can work on all that before you start to populate the dashboard. Because

once you start to populate and you want to come back into working on fundamentals, then it gets a bit clumsy.

Right? So dev is done. I worked with him a bit to make the background look exactly like the way I wanted it. It was

just about giving him the example that I gave to him, iterating it to make sure it works. I also made him work a bit on

it works. I also made him work a bit on the header. So for instance, I told him,

the header. So for instance, I told him, hey, change the text of the hemis from blue to white and also increase the font in the header sections because I found the fonts to be too small. Now the

fundamentals is done. The layout looks good. So we can start populating the

good. So we can start populating the dashboard. Now before we do that very

dashboard. Now before we do that very very important from my experience with building mission control dashboards sometimes when the agents are doing things they basically screw up and that you don't have any reference or you don't have any backups. So what I'm

going to tell dev is that hey on the dashboard please create a dedicated backup folder and then I'm going to tell him this is where I want the backup to be and then basically before he makes

any version or update to the dashboard he should make sure that he makes a backup before he makes the changes. Now,

this way whenever something screws up, you can always go back to the backup. I

built a dashboard for over days and then my agent just screwed something up when he was trying to change something and I completely lost it because there was no way to go back. And this is why I would highly recommend that you create this backup so that you can always revert to

an older backup if something screws up in the build. So, he's done. He created

a backup folder and also created all the dependencies that we needed. And you can see now on the dashboard we have version 1.1, which means that we're going to be seeing the versions. Now we're going to start populating the dashboard. We're

going to start with building the overview tab. Here also I'm giving the

overview tab. Here also I'm giving the agent exactly how I want the overview tab to look like and I'm also giving him a design orientation as to what I want to go for. I'm going to send this to the agent and then you're going to get a

visual and then we're going to see exactly what I was asking of him from the visual. All right. So dev says he's

the visual. All right. So dev says he's done. Now I'm curious. Let's see what he

done. Now I'm curious. Let's see what he builds. I'm going to make this a bit

builds. I'm going to make this a bit bigger and I'm going to reload the page.

Oo, this looks good. So you can see what actually happens here is that I told dev hey I want to have a radar that basically it's just a visual this counts through all the agents and then the most active agent should be extreme and then

here you can see that the most active agent is deaf and that is why he is at extreme and then on the right side we basically see our VPS health. We want to see how much of the CPU we are using. We

want to see how much of our RAM we are using. We also want to see how much of

using. We also want to see how much of the disk that we are using just so that if our VPS is getting bloated up or if our VPS is choking we can see this immediately. And then very important we

immediately. And then very important we can also see the size of the current hemis database so that we can trim this down if we see things are going overboard. And down here we basically

overboard. And down here we basically see integrity. We see how many successes

see integrity. We see how many successes that we have for what the agents are doing. We see the number of agents calls

doing. We see the number of agents calls that we have. These are the number of messages that we've sent so far. And

this is the amount of tokens that we have put in. And this is the size of the cache that we have over here. You can

also see the up time. You can see the task today. You can see the number of

task today. You can see the number of errors. And then you can also see the

errors. And then you can also see the sessions. Now, what I would personally

sessions. Now, what I would personally like to have is that I find that the sections are a bit too stacked on each.

I would like to have some breathing space between the sections. So, what I'm going to tell them is that hey, I like the overall look, but I feel that the sections are a bit too packed together.

So he should add some more vertical spacing like either using bottom margin or some pattern so that the layout has some more breathing room. I'm going to send this to him. All right. So dev says he's done. Let me refresh this and see

he's done. Let me refresh this and see if I have All right. So I think this looks good. So we see that I have more

looks good. So we see that I have more pattern between the sections. Now

another thing that I would like to have from a visual perspective is that just to make it a bit more cleaner. I would

prefer that instead of writing 1 2,222 I'd rather like to have shorter text like 1.2 m. And then for messages for instance 1.3K and stuff like that. And

also I think from a visual design personally I would like to have the text here having the same color as the borders at the top. So I'm going to copy this prompt and I'm going to give it to death to implement this as well. So

looks like he's done. Now let's refresh the dashboard to see if that is also working. Perfect. So now it looks more

working. Perfect. So now it looks more like what I would like to have. I have

the colors matching the section at the top. Also rather than having 10 long

top. Also rather than having 10 long text, we have this truncated into the version that we would like to have. Now

in the next stage, we're going to work on the agents tab. The idea that I have is that I would like to have different cards at the top here for all the agents and I would also like to have a general overview of how the agents are working, the tasks they have performed and stuff

like that. Now for that, I'm going to

like that. Now for that, I'm going to copy this prompts here and I'm going to give it to def and then I'm going to explain the prompts whilst he's working.

So what I'm technically doing is that I'm telling hey build a full agent tab.

I want to have a large display of a text the collective just to imply that I have a collection of agents. And I'm also telling him basically what a design there should look like. And I'm saying I

want to have five agents cards in a row so that we have all the five cards listed together. And in each card I'm

listed together. And in each card I'm basically explaining that I want to have the agents name. I want to have a twoline description of the latest thing that they did. I want to have their 7-day activity. I want to have a sort of

7-day activity. I want to have a sort of a bad chart to basically see a visual of on which day they were more active and stuff like that. And this is the design orientation that I'm giving to dev to build this. Now let's look at what he's

build this. Now let's look at what he's going to make out of this. All right, so he's done. Let's see what he built. I'm

he's done. Let's see what he built. I'm

going to reload the page. I'm going to go to agent. Oops. This looks good. So

the only thing that I see here is that it appears the content bleeds out of the section here. So I'm going to tell him

section here. So I'm going to tell him to fix this. But otherwise, everything looks good. We have the five agents

looks good. We have the five agents card. It's written clearly that

card. It's written clearly that orchestrator is on telegram, scout is on discord and all the other agents are on discord. We see their role, their

discord. We see their role, their serving this activity, but just we also see the model that they used but just that we see that the content bleeds out of the card. I'm going to tell him the section already looks good visually. The

only issue is that some of the content like here are trimmed off. So I want him to fix the layout so that everything looks quite nicely in the agent cuts.

And I'm going to click on send. All

right. So dev is done. Now let's reload the page to see if the content in the cards are fixed. I'm going to go to the agent tab. Perfect. So now you see that

agent tab. Perfect. So now you see that everything looks good. Every content is well contained in the respective cards.

Now the next thing that I'm going to do is I want to add sort of a visual overview of like a task statistics so that we can see from one go how the agents are doing, which agent is most active, a pie chart to see how the

activity is and stuff like that. And to

do that, I'm going to copy this prompts here. I'm going to call this the agent

here. I'm going to call this the agent statistics and I'm going to send this to dev to implement this design as well.

All right, so he's done. So when I reload the dashboard, you see that we have this agent statistics tab where we can see the number of attacks that have been performed today, what has been performed this week, the success rate, the most active agent and stuff like

that. But when I got here, what I

that. But when I got here, what I realized is that it appears that the agents are actually not logging when they work. For instance, dev has been

they work. For instance, dev has been working for a while, but you can see that he is not logging. The last

activity that we see from dev by filter to dev was 1 hour ago, which is not true because I've been building the dashboard with him the whole time. Now, to fix this, we're going to give the agents this prompt. I'm telling, hey, store

this prompt. I'm telling, hey, store this in your long-term memory. After

every task, I want you to basically log every activity that you performed. And

I'm also giving the agent an instruction on how they should do this because without this, we wouldn't be able to monitor what the agents are technically doing. So I'm going to send this to dev

doing. So I'm going to send this to dev and I'm basically going to see how he reacts to this. All right. So you see that it's saved to his memory and immediately he did that. I can see from the dashboard that save the agent login

memory. So basically he's logging. Now

memory. So basically he's logging. Now

I'm going to send this to all the agents just to be sure that the agents are logging the activity because without this we wouldn't be able to basically track what they have been doing and it also makes the dashboard more interactive. So now you can see that I

interactive. So now you can see that I have four agents active and when I come down here into the agent locks when I click on all you can see that I now see that reach has logged reach has saved this to his long-term memory scout saved

and all the agents are now saving and this looks very good. So now we can proceed to the next step. So I'm going to tell dev awesome so now are you ready to continue building the dashboard? So

in the next stage now we have the agents tab, we have the overview tab. The next

thing that we're going to do is we're going to build the task tab. Now the

idea here that I have is that you have your own sort of personalized tax tracker where you can basically move stuffs into to-do in progress and completed so that you have your own sort of productivity sort of organizer. I

just want to show you how you can build this because I use this a lot to basically organize the things that I have to do into to-do and stuff like that. And I find this to be very

that. And I find this to be very helpful. So I'm going to give dev the

helpful. So I'm going to give dev the prompt to build this task tab. So dev is done. Now let's go to the task tab and

done. Now let's go to the task tab and let's reload the page. Then when I come to task, beautiful. So now you see that you have your own sort of canban dashboard where you can have stuffs into pending in progress and done. You can

basically drag stuffs in between when you are working on this progress. Now

the next stage that what we're going to work on is we're going to work on sort of a schedule tab. The idea of the schedule tab is that you can basically have an overview of the chrome jobs that are running in your hemis setup. You can

see what is upcoming, what is next, and stuff like that. So to do that I'm going to copy this prompts here and I'm going to tell dev to also populate this schedule tab. Now this is basically

schedule tab. Now this is basically supposed to give us a visual overview of all the operations that are going on in our hemis setup so that we can see what the next task is what the agents are going to do in terms of chrome jobs and

stuff like that. Think of it like a chrome calendar more or less. Now chrome

jobs are things that run periodically every day or every hour or every week or every month depending on how you define it. And an example of this, for

it. And an example of this, for instance, I can tell scouts that, hey, every morning I would like you to send me news around AI and automation for my

YouTube channel. And this is going to be

YouTube channel. And this is going to be a crone job because we're telling Scout that, hey, I want you to send this every day. You can also change this to

day. You can also change this to whatever you want it to be. And then

typically, it's going to ask you at what time do you want to send it. So if I say that, send it to me every day at 8:00 a.m., then scout is going to make sure

a.m., then scout is going to make sure that every day at that time he's going to send it. And this is exactly what he's asking. At what time do you want me

he's asking. At what time do you want me to send this? And I'm going to say send this at 8:00 a.m. local time. So that

means at 8:00 a.m. every day, Scout is going to perform this task and he's going to send me this report. And this

is what is called a Chrome job. And then

you can see that he says, "Hey, I'm going to send this to you." And you can see that he says, "Hey, this is scheduled. We have a daily AI and

scheduled. We have a daily AI and automation news for my YouTube channel.

I'm going to send it at this time." And

the delivery is going to be back here in his channel. And the first run is

his channel. And the first run is currently scheduled for tomorrow at 8:00 a.m. And then here as well, you see the

a.m. And then here as well, you see the job ID. Now, I'm going to go back to Dev

job ID. Now, I'm going to go back to Dev and see what Dev is building with the schedule tab. So, I can see that he's

schedule tab. So, I can see that he's still working. So, let's wait till he's

still working. So, let's wait till he's completed. Hey, if you find this video

completed. Hey, if you find this video helpful, please like and subscribe to my channel to see more like this so that the video also gets pushed out to the communities out there. You can see a lot of work goes into building these

tutorials, building the prompts, testing to make sure that things are working.

And I've tested this over 20 times to make sure that things are working before I present it to you. If you're finding any value in the video, I would highly appreciate if you like this video and subscribe to my channel to see more like this. It doesn't cost you anything

this. It doesn't cost you anything extra, but helps the YouTube algorithm to push the video so I can keep coming up with content like this to you. If you

want to help my work, you can also come to my website, computer mechanic.com, and you can also give me a tip, buy me a coffee, if you want to support the work that I'm doing. It's not mandatory, but hey, this goes to help the work that I'm

building. And what I find very nice

building. And what I find very nice whilst dev is working by the way with Hermes that I wasn't there with open claw is that you see that he has his own inbuilt browser like a headless browser that it basically checks whatever it

does. It can click, it can take

does. It can click, it can take screenshots. So it doesn't really render

screenshots. So it doesn't really render what it's building by code alone but it also has the visual to basically see exactly what it's building. So like if things are not working, it fixes it automatically. And this is one of the

automatically. And this is one of the beauty of Hemis that I'm really finding it tempting that I might switch from open cloth to Hemis because Hemis is really on a different level. All right,

so dev is done. You can see that we have an overview of the total jobs that are running periodically, the ones that are system jobs, the ones that are Hemis related jobs and you can also see for instance when the next run is going to

be. Now when you come into the monthly

be. Now when you come into the monthly section for instance, you can see that the job that we set where we want to clean the database every month is actually already here. And what I do not

see surprisingly is that the scouts, we give scouts this chrome job that he's going to run every day, but we do not see this on the schedule tab. So I'm

telling that, hey, this looks good, but somehow I don't see the chrome jobs that the agents themselves I'm asking them to run. And I'm giving him an example of

run. And I'm giving him an example of the one that Scout is actually doing that I do not see here. All right, so dev is done. Now let's reload the page.

And then hopefully we should now see the Chrome job from Scout. So you can see this is now appears here. Scout is going to run this every day. And this gives you a very nice overview of exactly the current jobs that are running. Now, the

last tab that we're going to work on is the content tab. The idea here is that periodically your agents are going to be generating long form documents, right?

And it gets a bit clumsy if the documents are stored here and there. So,

what we want to do is we want to create a dedicated folder under hemis called content. And this is where basically the

content. And this is where basically the agents are going to store their long form documents. And every agent is going

form documents. And every agent is going to have their own folder where they're going to store those content. And then

from the dashboard, we're going to be loading all the documents that are stored in there. And this gives us a very nice visual overview because otherwise if the agent writes content and they just dump it into the chat, it gets too clumsy. But this way, you're

going to have a very nicely organized space where you can see all the things that the agents are writing and gives you a very nice overview of what is going on. The advantage here as well,

going on. The advantage here as well, which I also personally even do, is that I also sometimes create my own content.

You can write your own content here and then save it as well as your own document space. All right, so dev is

document space. All right, so dev is done. So you can see that he added a

done. So you can see that he added a screenshot of exactly what he built, but we're going to see this ourselves. And

he's also doing a pre-mpaction because he realized the context window is getting too large. Now let me reload the page and let's see how the content tab looks. Now this looks good. So you can

looks. Now this looks good. So you can see that we have our own library where we're going to see all the contents that the agents are writing. We can also filter this by the agents themselves.

And this looks good. Now we have the content tab ready but the agents do not know that this exists and the agents also do not know how we want them to write content. Now so we're going to

write content. Now so we're going to copy this prompt here and we're going to send it to all the agents. We're going

to tell them hey whenever you are producing a long form document being it's an article research script and stuff like that I want you to save it in your dedicated agents folder and not return this in the chat window because

whenever agents past something in the chat the context window gets bloated and it's really a mess and that is why this is very helpful and then we also telling the agents what the folder structure looks like and we are giving them some

instructions on how to do this. So, we

want them to save the file starting with the date and then the article and also we want the first line to include the title. And basically here, I'm just

title. And basically here, I'm just giving them some formatting orientation of how I want things to be. So, I'm

going to copy this prompt and I'm going to send this to all the agents that we have. And additionally, I also told the

have. And additionally, I also told the agents to sort of do a smoke test where I told them to do a short test document so that I can confirm that it's working correctly. All right. So, the agents

correctly. All right. So, the agents confirmed that they are done. Let me go to the content tab. Let me reload the page. And let me go back to content. And

page. And let me go back to content. And

then now you can see that we have the test content. So you can see this was

test content. So you can see this was written by Scribe. This was written by reach dev and stuff like that. And this

gives you a very nice view of long form documents that the agents write. So now

we're getting close to the end of the tutorial. In the next stage, I'm going

tutorial. In the next stage, I'm going to show you two ways that you can access the mission control dashboard without having to put in your password every time. The first method that I'm going to

time. The first method that I'm going to teach you is what we're going to do using what is called SSH keeper. The

second method that I'm going to teach you is we're going to use something that is called tail skill. What I typically recommend in my case is I use the second method which is with tail skill because with tail skill you can access the

dashboard wherever you are on whatever PC so long as your tail skill account is on that PC and you can even access this on your mobile phone. So on my mobile phone I'm able to access my mission control dashboard. So I always don't

control dashboard. So I always don't have to be sitting on my PC to be able to access the dashboard. But with the first method which is the S8 tunnling this only works when you are using your PC and you can't use this method on your

mobile phone. But I'm going to walk you

mobile phone. But I'm going to walk you through all the two processes and then you can decide which one works best for you just to demonstrate what I mean.

Currently whenever we want to connect to the mission control dashboard we have to do this through SJ tunneling where we put in this command here the port and all the IP addresses to connect to the dashboard and after that we typically

press enter. Now in the next stage we

press enter. Now in the next stage we are always prompted to put in our password and only when we put in the password the SS tunneling get activated and we can then access the dashboard.

But it's sometimes a bit annoying when you just want to have access to the dashboard and you always have to go through this long process to activate a sit tunneling put in your password every time. And if you have a very complex

time. And if you have a very complex password like mine, this can get pretty annoying. And that is why I'm going to

annoying. And that is why I'm going to show you an automated way that we can do this where we're going to run an auto start script which you can just double click on your desktop and it automatically locks you in into the VPS

server and you can access the dashboard without having to type in your password every time. And to do so, we're going to

every time. And to do so, we're going to generate what is called an SSH key pair.

one of the key is going to stay locally on your PC and we're going to send the other key or the second key to the VPS.

So whenever your PC is trying to connect to the VPS, if the key that is stored on your PC is compatible with a key that is stored on the VPS, then your VPS would allow your PC to connect to it without

you having to put in a password. But if

these two keys are not the same, then you wouldn't be able to connect to the VPS server. And that is why this is one

VPS server. And that is why this is one of the secured ways that you can connect to the VPS server. And here as well, I'm going to provide all the commands that we're going to use. So you can just copy paste this and you'll be good to go. Now

to get started, we're going to have to first check to see if you have any SSH keys stored locally on your PC. Now you

can check this by just using this command here. I'm going to paste in this

command here. I'm going to paste in this command here. And this is basically

command here. And this is basically going to query to see if your PC has any SSH key pair already stored on it. So in

my case, you see that I have false. So

that means that I do not have any SSH key stored on my PC. If in your case you have true then it means that you have an SSH key pair. So you can just send this key directly to your VPS. Because I do

not have any key stored here. I'm going

to generate a key. And to generate a key we're going to use this command here.

SSH key gen minus t all this. And then

you're going to put this in quotation.

In my case I call this agent dashboard.

But you can also replace this agent dashboard text with anything that you want. And after that just press enter.

want. And after that just press enter.

Now it's going to ask you where you want to save the key. By default, SSH key pairs on Windows are always stored in this location. So, I'm going to leave

this location. So, I'm going to leave this at the default, and I'm just going to press enter. In the next stage, you're going to be asked if you want to add a password to this SSH keys that have been generated. In my case, this is locally on my PC. I'm not going to be

sending this anywhere, so I'm not going to put in any password. I'm just going to skip this by pressing enter. And it's

going to ask me again to repeat the password, but because I didn't use any password, you can just press enter to skip that option again. And then you can see that an SSH key pair has been generated for us. Now if you go into

this file path in your VPS, you would see that we have two things that have been generated. So this is the first key

been generated. So this is the first key that was generated and this is the second key that was generated for us.

Now we're going to keep one of these keys on the PC and we're going to send the second key pair to the VPS. And

after that, we would be able to connect to the VPS without having to put in our password. Now, let me clear the screen

password. Now, let me clear the screen here on PowerShell to make it a bit cleaner for you guys to see what I'm doing. So, what we're going to do now is

doing. So, what we're going to do now is we're going to copy this SSH key that we generated and we're going to send it into our VPS. And to do so, we're going to use this command here, SSH copy ID

root. And then you basically going to

root. And then you basically going to just replace this part here with the IP address of your VPS server. And then

after that, press enter. My bad, you're not going to have this. I was using the command for Mac OS or Linos OS. That is

why Windows didn't recognize this command. The right command for Windows

command. The right command for Windows is the one here. I'm just going to copy this command here and I'm going to paste this in notes editor. And this is the command that we're going to use. Now

over here, what you're going to have to do is to replace this section here with the IP address of your VPS server. I'm

going to paste this in here. And after

that, you should technically be good to go. If you're using a Mac operating

go. If you're using a Mac operating system, the command that you would use is what we use initially. But this as well, I have also outlined in the guidelines. So you would find the right

guidelines. So you would find the right command depending on your operating system that you're using. So I'm going to paste in this command. Now, because

this is a script that is going to send something to our VPS, Windows PowerShell is going to warn you if you're sure you want to paste this command. I'm going to say paste anyway. And after that, we're just going to press enter. Now,

PowerShell is going to make sure that you are the owner of the VPS before it sends this key to your VPS. So, for the last time, we're going to have to put in the password of our VPS server. So, I'm

going to put in the password. And after

that, I'm going to press enter. And then

once it brings you back to this prompt here, it means that the SSH key has been sent to your VPS server. Now let's try this again. Let's try to connect to the

this again. Let's try to connect to the VPS server using the same SSH root IP address of the VPS server. And let's see if it works. If it works, then your VPS is not going to ask you to put in a

password anymore. So let's try this now.

password anymore. So let's try this now.

And now you can see that I'm connected to the contactable VPS server without me having to put in a password. Now this

means that the SSH key that we generated was correctly sent to the VPS. Now if I want to try the SH tunneling again to connect to the mission control dashboard, remember that we were using

the command SSH minus L the port and the IP address of the VPS server. This is

how we used to activate SSA tunneling to be able to access the dashboard. So you

can see that when I'm not running this, the mission control dashboard doesn't load because I haven't activated the SSA tunneling. Once I run this command now,

tunneling. Once I run this command now, it would run without asking me for a password because the SSH key is stored on my PC and the pair is stored on the VPS. And now you can see that the SE

VPS. And now you can see that the SE handling is connected without me having to put in the password. And if we reload the mission control dashboard now, you can see that the dashboard is working without me having to put a password. Now

that was just about setting up the password to allow us to be able to seamlessly connect to the VPS without having to put in a password. But the

best part is what we're going to do.

We're going to run an automated script that would allow us to be able to in one click activate this SA tunneling without us having to go through all the complex process with PowerShell. And this is how we're going to do this. Let me close the

PowerShell terminals that we have opened and let me open a new PowerShell terminal. Now, because this is a script

terminal. Now, because this is a script that we're going to run, we're going to have to tell PowerShell that, hey, we're going to be running some scripts and I want you to be able to run these for me automatically. And to do so, we're going

automatically. And to do so, we're going to use this command here. Just copy this command. come into PowerShell and paste

command. come into PowerShell and paste in this command and press enter. Now

once it brings you back to the prompt here, it means that this has been activated. So your PowerShell would

activated. So your PowerShell would allow scripts to be able to run automatically. In the next stage, just

automatically. In the next stage, just open notepad editor and then copy the script that I have here. Copy everything

from here and then we're going to paste in the script. Now what the script is just saying is that hey, I want to activate SSA tunneling on a particular port. So in my case, I'm going to have

port. So in my case, I'm going to have to just change this port from 45 68 because our mission control dashboard is hosted at the spot, not the port that I had in that tutorial. So just copy this

part here and then replace it. But what

I'm going to provide to you at the end of the day has a corrected port. So all

you're going to have to do is to just replace your VPS IP here. But first,

let's go through what the script is saying. The script is saying that hey, I

saying. The script is saying that hey, I want to run an automated script through SSA tunneling. And if there is any SSA

SSA tunneling. And if there is any SSA tunneling already running, please kill this process so that I will be able to access the SSA tunneling. And then here we just going to get a message that starting the SSA tunneling to the agent

OS. And you're going to get a message

OS. And you're going to get a message the dashboard will be available at local host here. Again, let me change this

host here. Again, let me change this because it's a different port than what our dashboard uses. And then the window would just stay open so that you can use the dashboard. And then very important

the dashboard. And then very important at this step here, replace this part here with the IP address of your VPS server. So, like I said, with the

server. So, like I said, with the prompts that I'm going to give you at the end of the day, I'm going to change all the ports. And all you're going to have to do is to just put in your IP address of your VPS server, and you will be good to go. And after that, just come

to file, then come to save as, and then just navigate to the folder where you want to save this script. In my case, I'm going to save this on my desktop.

And I'm just going to call this MSOS.

But very, very important, save the file with the extension. So, whatever file name that you give to it, end with PS.

Now this is going to just tell Windows that hey this is a script that is going to be run by PowerShell and after that just click on save and then once you are done you can close the script. So what

I'm going to do is I'm going to navigate to the folder where I saved this script.

So now I'm in that folder and you can see that the script has already identified that is a script that would be run with PowerShell. Now what I'm going to do just to make sure that this is going to work I'm going to reload the mission control dashboard. You can see

that currently I can't access the mission control dashboard. But

immediately I come to the script, right click on the script and then choose run with PowerShell. You now see that it

with PowerShell. You now see that it tells you hey keep this window opened whilst you are using the dashboard.

Technically the SSA tunneling is already activated. And then when you come to

activated. And then when you come to your browser and you reload the dashboard you see that hoola it opens the mission control dashboard for you.

And you can see that this is very easy way to access the dashboard. Whenever

you are done just press come to the terminal. You can close the PowerShell

terminal. You can close the PowerShell application or you can press Ctrl C and then it basically kills the application and then you see that the dashboard doesn't work anymore. Whenever you want to open the dashboard just right click

and come to run with PowerShell it opens the SA tunneling for you and then when you reload the dashboard you can see that now you are connected to the dashboard. So that is one of the ways

dashboard. So that is one of the ways that we can connect to the dashboard using SSH tunneling. But there is an even better way that you can connect to the dashboard using what is called tail skill. Now, Tailskill sort of sets its

skill. Now, Tailskill sort of sets its own VPN on your PC. The advantage here is that wherever you are, so long as you have Tails skill installed and connected to that PC, b your desktop here, b it to

your laptop, b your mobile phone, you can always access the mission control dashboard. And this is what I personally

dashboard. And this is what I personally use because I do have several PCs that I'm working with. When I'm on my Mac, I can access the mission control dashboard. When I'm on my Windows, I can

dashboard. When I'm on my Windows, I can access it. When I'm on my mobile phone,

access it. When I'm on my mobile phone, I can also access the mission control dashboard. So long as I have Tail Skill

dashboard. So long as I have Tail Skill installed there. And I'm going to walk

installed there. And I'm going to walk you through how we can set up this as well so that you can connect to the dashboard very easily. Now with tail skill whenever you activate tail skill you don't have to go through all the SSA tunneling put in your password and all

those complicated things that we did.

It's going to be so seamless to be able to connect to the dashboard. Now to do so the first thing that we're going to have to do is to just come to your dashboard. Just come to Google and just

dashboard. Just come to Google and just set tail skill. And once you are here what we're going to do we're going to create a free account. You would not have to pay anything for what we're going to do. Just come to get started

where it says it's free and then you can just basically create an account and then once you get here you're going to be walked through a setup for tails skill to be able to understand exactly what you want. So what is your primary reason for using tail skill just choose

anything that you want application testing what is your role. I'm just

going to go with it and I'm going to go to next to add your first device. Now at

this stage we're going to download tail skill. So you can click on download tail

skill. So you can click on download tail skill for Windows and I'm just going to download tail skill for Windows and then when you come back here just click on skip this introduction as well. Now you

can also always download the tail skill application by just coming to their website and coming to download and you can download the application as well.

But if you're starting this for the first time they ask you this in the process. Now tail skill is installed on

process. Now tail skill is installed on my PC. Now we're going to run the tail

my PC. Now we're going to run the tail skill application. Just going to click

skill application. Just going to click on agree. I'm going to click on install.

on agree. I'm going to click on install.

I'm going to click on yes. And the

installer is technically running. So now

we have tail skill installed on our PC.

The UI for tail skill on Windows is very very horrible. So it's not an

very horrible. So it's not an application that you can double click to open as an application. You always have to work in the taskbar here. So just

come to the tax bar here and this icon with the nine dots is basically tail skill. Just right click on it and then

skill. Just right click on it and then if you do not have an account you can see that it tells you that it needs authentication. So just click on need

authentication. So just click on need authentication here or you can also basically right click here and come to login. Then it's going to bring you to a

login. Then it's going to bring you to a page where you're going to have to login into your Tailskill account. Now once

you're logged in, it's going to bring you to this page here where it's going to tell you, hey, you're going to connect your home desktop to your Tailskill account. Do you want to

Tailskill account. Do you want to connect this? I'm just going to say

connect this? I'm just going to say connect. And then after that, it tells

connect. And then after that, it tells you login is successful. So technically,

this PC that I'm using here is now connected to my Tailskill account. Now

the next stage is very important. We're

going to have to tell dev that hey, we want to connect the mission control dashboard that we just set up to tail skill. So he should do the setup for us.

skill. So he should do the setup for us.

So that after that we can just connect the VPS to our Tailskill account and we will be able to access the VPS through Tailskill. And to do that we're going to

Tailskill. And to do that we're going to tell him this command here. Please set

up Tails skill on the VPS so I can securely connect to the dashboard from anywhere both on my PC and on my mobile.

So I'm going to tell him to please install the Tailscale for us on the VPS start authentication process and he would provide us with a link that we would use to authenticate to connect the VPS to our Tailscale account. So you can

see that he brings a command that he wants to try to install tails scale for us. I'm going to say that always allow

us. I'm going to say that always allow so that he can proceed with the installation. So dev is asking for

installation. So dev is asking for another approval. I'm going to allow

another approval. I'm going to allow this as well. All right. So dev is done.

Tails scale is installed and we just have to approve this. Now to approve this for this VPS, we're going to use this URL here. So just copy the URL.

Come to your browser. We're going to paste in the URL and after that just press enter. Then we're going to login

press enter. Then we're going to login into our TKL account. And then very important tails is going to tell you hey you want to connect your account to this particular device. In my case this

particular device. In my case this device is the ID of my VPS. And if

you're sure this is the right one just click on connect. And then you see that you are told login successful. So now we can come back to discord and then dev says that once we are done with authentication we should just reply

approved and he would verify that this is working. So I'm just going to type in

is working. So I'm just going to type in approved and then he would get the tail skill IP for us to be able to work with skill. All right. So dev says he's done.

skill. All right. So dev says he's done.

Now let's try to see if this is working.

So he gives us two ways that we can access the dashboard. One is using this URL here which we can copy. First let me make sure that this is let me kill any process that is so here I have SSA tunneling running. Let me cancel the SA

tunneling running. Let me cancel the SA tunnel. Let me close all the PowerShell

tunnel. Let me close all the PowerShell just to be sure that it's tail skill working. Now let me reload the

working. Now let me reload the dashboard. So now it shouldn't work

dashboard. So now it shouldn't work because I have closed the PowerShell.

Now let's put in the URL that dev just gave to us. If it works then it means that the tail skill is working correctly. When I run this, hooray. You

correctly. When I run this, hooray. You

see that now we can access a dashboard.

And now we are not doing this through using PowerShell. We are doing this

using PowerShell. We are doing this through SSH handling. And when you come back here, he also gives you another URL here which you can use for instance this one here. When I copy this, so this one

one here. When I copy this, so this one basically has the IP and a port. But you

can also try something like this which gives you a full URL. And this can also work as well. When I put in this command, you see that it also opens the URL. Now the advantage here is that

URL. Now the advantage here is that whenever you have tail skill installed on your PC. When I come here and for instance I disconnect tail skill you see that the dashboard would not work anymore. So it only works when you have

anymore. So it only works when you have tail skill running. So when I right click on here and I say connect and I reload the dashboard. Now you see that we can access the dashboard. Now my tip for you would be that just bookmark this

page somewhere. So then whenever you

page somewhere. So then whenever you want to access the dashboard you can just come to the bookmark click on it and it would basically open that dashboard for you. Another tip for you, Tailskill sets its own VPN on your PC.

So please make sure that whenever that you are using Tailskill, you do not have any VPN application running because if a VPN is running, then Tails skill will not be able to run at the same time. So

only one VPN service has to be running at a time for you to be able to use Tailskill. I know this was probably a

Tailskill. I know this was probably a lengthy tutorial, but you can see that I really covered everything that you guys need to know about Hermes and the mission control dashboard and also how to connect to the mission control dashboard. If you do have any

dashboard. If you do have any difficulties in setting this up or if you do have any questions during the setup, just hit me up. Drop this in the comment section and I'm very happy to respond to you. If you want one-on-one

support with any personal issues that you were having during the setup, you can also just book a call with me, come to my website, book a session with me oneonone, and then we can have a look at it together. And with that, I say thanks

it together. And with that, I say thanks for watching this video and see you in the next one. Bye.