Automate ANY task using ChatGPT! (with GPT actions feature)

AI Foundations
9 Sept 202426:55

Summary

TLDRDieses Video zeigt, wie man Chat GPT nutzt, um Aufgaben zu automatisieren und relevante Informationen aus verschiedenen Quellen wie WordPress, Google Docs, Google Sheets, Social Media oder E-Mails schnell abzurufen. Es wird ein umfassendes Handbuch und eine schrittweise Checkliste gegeben, um die Produktivität sowohl im privaten als auch im beruflichen Leben zu steigern. Es wird auch die Verwendung von Webhooks und die Erstellung von Automations in Make.com behandelt, um die Interaktion mit externen Anwendungen zu erleichtern.

Takeaways

  • 🤖 Chat GPT ist ein mächtiges Werkzeug zur Automatisierung von Aufgaben, das richtig genutzt Zeit sparen kann.
  • 🔄 Um Zeit zu sparen, sollte man vermeiden, ständig zwischen Chat GPT und anderen Anwendungen zu wechseln.
  • 📑 Der Videoinhalt zeigt, wie man in Chat GPT bleiben und relevante Informationen aus verschiedenen Quellen abrufen kann.
  • 💌 Mit Chat GPT kann man Informationen senden, etwa per E-Mail oder in Google Sheets, ohne die Anwendung zu verlassen.
  • 🔗 GET-Anfragen werden verwendet, um Informationen von externen Anwendungen abzurufen, während POST-Anfragen verwendet werden, um Daten an externe Anwendungen zu senden.
  • 📈 Der Videolehrer bietet umfassende Ressourcen und eine schrittweise Checkliste für die erfolgreiche Automatisierung von Aufgaben.
  • 👥 Die AI Foundations Community wurde gegründet, um Menschen effizienter und produktiver zu machen, sowohl im privaten als auch im beruflichen Leben.
  • 🛠️ Automatische Prozesse können durch das Erstellen von Webhooks und die Verwendung von make.com-Automatisierungen eingerichtet werden.
  • 📝 Es wird ein Beispiel gegeben, wie man mit GET-Anfragen kürzliche Artikel sammelt und mit Chat GPT LinkedIn-Posts erstellt.
  • 🔧 Die Erstellung eines GPT mit spezifischen Anweisungen und Konversationsstartern wird detailliert erläutert, um die Automatisierung zu optimieren.
  • 🔄 Ein GET-Request-Beispiel zeigt, wie man seine E-Mails zählt, indem man eine Anfrage an Gmail sendet und die Anzahl und den Inhalt der E-Mails erhält.

Q & A

  • Was ist das Hauptthema des Videos?

    -Das Hauptthema des Videos ist, wie man die Effizienz und Produktivität durch die Nutzung von AI und Automatisierungstools wie GPT steigern kann.

  • Was verspricht der Sprecher im Video, den Zuschauern zu zeigen?

    -Der Sprecher verspricht, wie man in GPT bleiben und relevante Informationen aus verschiedenen Quellen wie WordPress, Google Docs, Google Sheets, Social Media und E-Mails abrufen kann, ohne die Seite zu verlassen.

  • Was ist die AI Foundations Community, die im Video erwähnt wird?

    -Die AI Foundations Community ist eine Gruppe, die dazu geschaffen wurde, um Menschen dabei zu unterstützen, effizienter und produktiver zu werden, indem sie AI in ihrem persönlichen oder beruflichen Leben nutzen.

  • Was sind GET und POST Anfragen und wie werden sie im Video erklärt?

    -GET-Anfragen werden verwendet, um Informationen von anderen Quellen abzurufen (dargestellt durch eine rote Linie), während POST-Anfragen verwendet werden, um Informationen auf andere Quellen zu posten (dargestellt durch eine blaue Linie).

  • Wie verwendet der Sprecher GPT, um seine LinkedIn-Posts zu automatisieren?

    -Der Sprecher verwendet GPT, um Nachrichtenartikel zu aggregieren, die er aus verschiedenen Quellen liest, und dann verwendet er GPT, um automatisch LinkedIn-Posts basierend auf dem relevanten Inhalt der Nachrichtenfeeds zu erstellen.

  • Was ist der Zweck des 'Schema Ninja', der im Video erwähnt wird?

    -Der 'Schema Ninja' ist ein von dem Sprecher erstelltes GPT, das dazu verwendet wird, um Schritt für Schritt Anleitungen für die Erstellung von Webhooks und Automationen in make.com zu geben.

  • Wie wird in make.com eine Automation erstellt, wie im Video gezeigt?

    -In make.com wird eine Automation erstellt, indem man zuerst einen Webhook erstellt, dann die Einstellungen ändert, um GET oder POST Anfragen zu ermöglichen, und dann verschiedene Module hinzufügt, um mit verschiedenen Apps wie Facebook oder Instagram zu interagieren.

  • Was ist der Vorteil der Verwendung von make.com für die Automatisierung?

    -Der Vorteil der Verwendung von make.com für die Automatisierung liegt darin, dass es Zeit sparen kann, indem es eine komplexe Abfolge von Schritten in einer einfachen und schnellen Weise ermöglicht.

  • Wie wird im Video gezeigt, wie man einen POST Request von GPT aus sendet?

    -Im Video wird gezeigt, dass man einen POST Request senden kann, indem man die Anweisungen in GPT ändert, um die Informationen im JSON-Format zu strukturieren, das für die Kommunikation mit dem Webhook erforderlich ist.

  • Was ist der letzte Schritt im Prozess der Automatisierung, wie im Video beschrieben?

    -Der letzte Schritt im Prozess der Automatisierung ist der Test und die Anpassung der erstellten Automation, um sicherzustellen, dass sie korrekt funktioniert und die gewünschten Daten abruft oder sendet.

Outlines

00:00

🤖 Chat GPT für Automatisierung

Dieser Absatz betont die Fähigkeit von Chat GPT, Aufgaben zu automatisieren, wenn man es richtig verwendet. Es wird erwähnt, dass viele Menschen Zeit verschwenden, indem sie Inhalte in und aus Chat GPT kopieren und einfügen. Der Sprecher will zeigen, wie man in Chat GPT bleiben und relevanten Inhalt aus verschiedenen Quellen wie WordPress, Google Docs, Google Sheets, sozialen Medien, E-Mails usw. abrufen kann, um die Produktivität zu steigern. Es wird auch erwähnt, dass die AI-Grundlagen-Community dazu beiträgt, Menschen effizienter und produktiver zu machen, ob in persönlichem oder beruflichem Kontext.

05:01

🔍 Grundlagen der Automatisierung und Terminologie

Der Sprecher erklärt die Funktionsweise von Chat GPT im Hintergrund und verwendet bestimmte Begriffe wie GET- und POST-Anfragen. GET-Anfragen werden verwendet, um Informationen von externen Anwendungen abzurufen, während POST-Anfragen verwendet werden, um Informationen auf andere Anwendungen zu posten. Es wird ein Beispiel gegeben, wie man eine GET-Anfrage stellt, um die Anzahl neuer E-Mails zu erfahren, und wie eine POST-Anfrage verwendet wird, um eine Geschichte über einen Hund in Google Docs zu erstellen und zu posten.

10:02

📝 Schritt-für-Schritt-Anleitung zur Automatisierung

Der Sprecher führt durch die Schritte, um eine GPT zu erstellen, die LinkedIn-Beiträge für ihn postet. Dies beginnt mit dem Erstellen einer GPT mit einem Namen, einer Beschreibung, einem Profilbild, Anweisungen und Konversationsstartern. Es wird auch ein Checkliste und eine Anleitung zur Erstellung der GPT bereitgestellt. Der Sprecher erklärt dann, wie man eine GET-Anfrage verwendet, um Nachrichtenartikel zu aggregieren und in Chat GPT zurückzuholen, um LinkedIn-Beiträge zu erstellen.

15:03

🔗 Erstellen von Webhooks und Automationen

In diesem Absatz wird erklärt, wie man eine Verbindung in Make.com herstellt, indem man einen Webhook erstellt und eine Make.com-Automatisierung einrichtet. Es wird ein GPT namens 'Schema Ninja' genannt, das verwendet wird, um die Schritte zur Erstellung der Anforderung zu erläutern. Der Sprecher führt durch die Erstellung eines neuen Szenarios, der Einstellungen für GET- und POST-Anfragen und wie man eine Automation erstellt, indem man Module hinzufügt und Apps wie Facebook oder Instagram verbindet.

20:03

📑 Erstellen von Webhook-Antworten und Datenzuordnung

Der Sprecher beschreibt die Bedeutung von Webhook-Antworten und wie man sie verwendet, um Daten zuzuordnen. Es wird erklärt, wie man die Automation abschließt, indem man eine Webhook-Antwort erstellt und Daten zuordnet. Es wird auch gezeigt, wie man die Schema-Ninja verwendet, um die GET-Anfrage zu erstellen, die für die Kommunikation mit der GPT erforderlich ist, um Informationen abzurufen.

25:05

🔄 Testen und Anpassen der Automation

In diesem letzten Absatz wird gezeigt, wie man die erstellte GET-Anfrage testet und wie man die Anweisungen anpasst, um die gewünschte Datenformate zu erhalten. Es wird auch erklärt, wie man die Schema-Ninja verwendet, um die POST-Anfrage hinzuzufügen und wie man die Automation anpasst, um LinkedIn-Beiträge zu erstellen und zu posten. Der Sprecher betont die Bedeutung der Dynamik und Anpassungsfähigkeit von Automationen und wie man sie effektiv nutzen kann.

🚀 Abschluss und Aufruf zur Community

Der Sprecher fasst die Automation zusammen und hebt hervor, wie man die fertige Automation testet und wie man die LinkedIn-Beiträge veröffentlicht. Es wird ein Aufruf an die Zuschauer gerichtet, der AI-Grundlagen-Community beizutreten, um mehr über die Verwendung von KI in verschiedenen Branchen zu erfahren und mehr Effizienz zu erlangen. Der Sprecher bittet um Abonnements und Likes und kündigt an, dass er weitere ähnliche Videos produzieren wird, falls gewünscht.

Mindmap

Keywords

💡Automatisieren

Automatisieren bezieht sich auf den Prozess der Erstellung von Systemen oder Prozessen, die alleine funktionieren können, um Zeit und Mühe zu sparen. Im Video wird gezeigt, wie man mit Hilfe von AI und GPT Aufgaben wie das Schreiben von E-Mails oder das Aktualisieren von Google Tabellen automatisch ausführt. Dies wird durch die Verwendung von GET- und POST-Anfragen demonstriert, um Daten abzurufen oder zu senden.

💡GPT (Generative Pre-trained Transformer)

GPT ist ein generischer Begriff für künstliche Intelligenzen, die verwendet werden, um Texte zu erzeugen oder Aufgaben zu erleichtern. Im Kontext des Videos ist GPT eine Plattform, die verwendet wird, um mit verschiedenen Online-Diensten wie Google Docs oder WordPress zu interagieren und diese zu automatisieren.

💡GET-Anfrage

Eine GET-Anfrage ist ein Befehl im HTTP-Protokoll, der verwendet wird, um Daten von einem Server abzurufen. Im Video wird dies verwendet, um z.B. die Anzahl neuer E-Mails aus Gmail abzurufen oder Nachrichtenartikel zu holen, die später für LinkedIn-Posts verwendet werden können.

💡POST-Anfrage

Eine POST-Anfrage ist ein weiterer Befehl im HTTP-Protokoll, der verwendet wird, um Informationen auf einem Server zu speichern oder zu senden. Im Video wird dies verwendet, um z.B. eine Geschichte zu einem Google-Dokument zu senden oder einen Beitrag auf LinkedIn zu veröffentlichen.

💡Webhook

Ein Webhook ist eine Methode, um Informationen zwischen Anwendungen zu übertragen, indem eine HTTP-Anforderung an eine URL gesendet wird, wenn eine bestimmte Aktion ausgeführt wird. Im Video wird ein Webhook verwendet, um eine Verbindung zwischen GPT und verschiedenen Online-Diensten herzustellen, um die Automatisierung zu ermöglichen.

💡AI-Grundlagen-Community

Die AI-Grundlagen-Community ist eine Gruppe oder ein Netzwerk, das im Video erwähnt wird und das darauf abzielt, Menschen zu unterstützen, AI in ihrer persönlichen oder beruflichen Lebensweise effektiver zu nutzen. Sie bietet Ressourcen und Anleitungen, wie im Video gezeigt, um die Effizienz und Produktivität zu steigern.

💡Schema Ninja

Schema Ninja ist ein Begriff, der im Video verwendet wird, um ein Hilfsprogramm oder eine Art von AI zu beschreiben, das dazu beiträgt, Webhooks und die damit verbundenen Schemas zu erstellen und zu verwalten. Es hilft beim Erstellen von GET- und POST-Anfragen und stellt sicher, dass die richtigen Daten abgerufen oder gesendet werden.

💡RSS-Feed

Ein RSS-Feed ist ein Webbasierte Format für Inhaltssyndikation. Im Video wird ein RSS-Feed verwendet, um Nachrichtenartikel von verschiedenen Websites zu aggregieren, die später von GPT gelesen und als LinkedIn-Posts verwendet werden können.

💡JSON

JSON (JavaScript Object Notation) ist ein轻量级的数据交换格式, der im Video verwendet wird, um die Struktur der Daten zu definieren, die in Webhook-Anfragen gesendet oder empfangen werden. Es hilft dabei, die Kommunikation zwischen verschiedenen Anwendungen zu ermöglichen.

💡Make.com

Make.com ist eine Plattform für die Automatisierung, die im Video erwähnt wird und verwendet wird, um Workflows zu erstellen, die verschiedene Online-Dienste wie Google Sheets oder Social-Media-Plattformen verbinden. Sie ermöglicht es Benutzern, durch Klicken auf einen Button in GPT Daten abzurufen oder auszuführen.

💡Router

Im Kontext des Videos ist ein Router ein Werkzeug in Make.com, das verwendet wird, um verschiedene Routen oder Pfade für Daten zu verwalten. Es hilft dabei, zu entscheiden, welche Aktionen ausgeführt werden sollen, basierend auf den eingehenden Anfragen, obwohl es GET- oder POST-Anfragen sind.

Highlights

Automating tasks with chat GPT can save time if used correctly.

People often waste time copying and pasting between applications.

The video will show how to stay within GPT to pull in information from various sources.

GPT can send information to other applications like email or Google Sheets.

AI should save time, and if it's not, the user might be doing something wrong.

The AI Foundations Community helps people become more efficient with AI.

Understand how GPT works under the hood to automate tasks effectively.

GET requests are for retrieving information, and POST requests are for sending information.

Example of a GET request: Fetching email count and content through GPT.

Example of a POST request: Writing a story and sending it to Google Docs.

A step-by-step checklist is provided to help with automation.

Creating a custom GPT for automating tasks like posting on LinkedIn.

Aggregating news articles using a GET request to create LinkedIn posts.

Creating a GPT with a name, description, profile picture, instructions, and conversation starters.

Using a web hook in Make.com to connect GPT with external applications.

Creating an automation in Make.com to aggregate information from external sources.

Using the Schema Ninja GPT to create and revise schemas for GET and POST requests.

Testing the GET request to ensure data is pulled into GPT correctly.

Revising the schema to include a POST request method for sending data.

Setting up dynamic mapping of data to the web hook response.

Creating a LinkedIn post draft within GPT using aggregated data.

Publishing the LinkedIn post directly from GPT using the POST request.

The AI Foundations Community offers resources for leveraging AI in various industries.

Transcripts

play00:00

chat GPT is amazing when it comes to

play00:01

automating tasks but only if you know

play00:03

how to use it right often times people

play00:05

are left copying and pasting stuff into

play00:07

chat GPT and outside of chat GPT and

play00:10

ultimately that wastes time so in this

play00:13

video I'm going to show you how you can

play00:14

stay in your GPT and pull in relevant

play00:17

information whether it be from WordPress

play00:20

Google Docs Google Sheets social media

play00:23

posts emails whatever it is you have the

play00:25

ability to stay in your GPT click a

play00:28

button and retrieve that information in

play00:30

seconds you also have the ability to

play00:32

send that information maybe you want to

play00:34

send an email maybe you want to update a

play00:35

row in your Google Sheets from your GPT

play00:38

without leaving your screen maybe you

play00:40

want to create a Google doc analyze a

play00:42

PDF whatever it is you have the ability

play00:45

to automate that and stay within your

play00:46

GPT the entire time and in this video

play00:48

I'm going to show you how it's done I'm

play00:50

going to be giving you so many resources

play00:51

I'm going to be giving you a

play00:52

step-by-step checklist and I'm just

play00:54

going to ultimately walk you through how

play00:56

to do this and how to see success when

play00:59

automating Tas ask in chat GPT listen AI

play01:02

should be saving you hours of your time

play01:05

right now if it's not then you're doing

play01:07

something wrong that's why the AI

play01:09

foundations Community was created it was

play01:11

created in order to enable you to be

play01:13

more efficient and more productive

play01:15

whether that be tasks in your personal

play01:17

life or tasks in your work life we show

play01:20

you the steps to get there if you want

play01:22

to learn how to leverage AI in your

play01:23

industry to become more efficient and

play01:25

more productive as a person then join

play01:27

this community using the link in the

play01:29

description or the top pend comment you

play01:31

will not regret it before we get into

play01:33

chat gbt and we start learning how to

play01:35

actually automate things let's

play01:36

understand real quick how it's working

play01:38

under the hood and some of the

play01:39

terminology I'm going to be using so the

play01:41

red line is representing getting

play01:43

information from an external application

play01:46

the blue line is representing performing

play01:48

actions on other applications okay so

play01:51

you start off in your GPT this is big

play01:53

chat GPT logo and you can either send a

play01:55

get or a post request a get request is

play01:58

the red line because you're getting

play02:00

information from other sources the post

play02:02

request is the Blue Line because you're

play02:04

posting information on other sources or

play02:06

to other sources so let me show you a

play02:08

quick example of a get request I might

play02:10

prompt to my GPT how many new emails do

play02:12

I have today and it will send that

play02:13

request to make and run through an

play02:15

automated Gmail process on make because

play02:17

it connects to Gmail and fetch my email

play02:20

contents and how many emails I have and

play02:22

the body content of those Etc and it

play02:24

might form back a response of you have

play02:26

five new emails an example of a post

play02:28

request might be me saying write a story

play02:30

about a dog and send it to my Google

play02:31

Docs when complete my GPT would take

play02:34

that request in the form of a post

play02:35

request which I'll show you how to set

play02:37

up contact make with the content that

play02:40

I've provided the story about the dog

play02:42

and send it to a new document within my

play02:44

Google database and what I've even

play02:46

provided for you is a step-by-step

play02:48

checklist down in the description that

play02:49

you can just follow along as we're doing

play02:51

this video together for everything you

play02:53

need I've even provided you a custom GPT

play02:56

for this system to help you when I'm not

play02:59

here to help on on the video so first

play03:01

you need to understand what you want

play03:02

automated and I'm sure you have some

play03:04

ideas right now uh even just by watching

play03:06

this first part of the video but let's

play03:08

say for me I don't have much time to

play03:11

create LinkedIn post but I read a lot of

play03:13

artificial intelligence news articles

play03:16

and I have specific websites I like

play03:17

going to so for me a good example of

play03:20

creating a GPT that is completely

play03:22

automated and post on LinkedIn for me

play03:24

would be step number one using a get

play03:26

request so I can aggregate all of my

play03:28

news articles and bring them back into

play03:31

chat GPT so that chat GPT can

play03:33

automatically create LinkedIn post for

play03:36

me based on the relevant content from

play03:38

the news feeds that I like so I'm

play03:41

pulling articles into my GPT creating

play03:43

posts using chat GPT that's why we would

play03:46

go to chat gbt for this and then I would

play03:48

send off those posts to my LinkedIn

play03:50

automatically that's what we're going to

play03:51

be building today so let's get right

play03:53

into it step number one what we need to

play03:54

do as you can see on our GPT automation

play03:57

checklist is we need to create our G PT

play04:00

so we need a name a description a

play04:02

profile picture instructions and

play04:04

conversation starters if you want a more

play04:06

in-depth guide on creating this version

play04:08

of your GPT like creating a very good

play04:10

GPT with amazing instructions I'll leave

play04:12

a link to that video in the upper right

play04:14

hand corner where I go much more in-

play04:16

depth on that so real quickly I'm going

play04:18

to create this GPT and then I'll report

play04:20

back to you once I'm done and kind of go

play04:22

through what I did what I'm going to do

play04:24

is in the upper right hand corner I'm

play04:25

going to go to my gpts and then I'm

play04:27

going to select create a GPT here I can

play04:30

give it a name description instructions

play04:31

conversation starters and a profile

play04:33

picture and again since I am automating

play04:36

my LinkedIn post I can give it a name of

play04:38

the LinkedIn master and then I can give

play04:40

a little description of what it does

play04:42

this is actually an image of the

play04:43

LinkedIn logo that I got generated for

play04:46

this on Gro which is another language

play04:48

model I can hit open and just like that

play04:50

we have a profile picture name

play04:51

description and now we need instructions

play04:53

and this is kind of going to be

play04:55

instructions for how your GPT needs to

play04:57

run all the actions you want it to

play04:58

complete and how it's going to contact

play05:00

your web hooks so for that I put you are

play05:03

a LinkedIn growth specialist now this is

play05:05

basics of prompt engineering if you

play05:07

don't know this yet then you definitely

play05:09

need to learn it before you get into

play05:10

fancy automations but I basically give

play05:12

it a roll and tell it its job I tell it

play05:15

what it's going to receive and then I

play05:17

say right here when I say fetch articles

play05:20

contact my web hook and get me the most

play05:22

recent articles on AI this is going to

play05:24

be a sendoff command to start the get

play05:26

request which is going to be important

play05:28

as you can see it says when I say fetch

play05:30

articles in order to make my job easier

play05:32

I can copy fetch articles and make a

play05:34

conversation starter so now I can hit

play05:36

this button and say fetch articles and

play05:38

it will automatically run through that

play05:40

process for me since I set it up like

play05:42

this in my instructions and then I

play05:44

basically tell it to receive the URL

play05:46

read the content and create a LinkedIn

play05:48

post and I give it a custom output

play05:50

format for each LinkedIn post now I can

play05:53

go back to our GPT automation checklist

play05:56

and step one is complete we have a name

play05:58

description profile picture instru

play05:59

instructions and conversation starters

play06:01

next this is where the real fun begins

play06:03

we need to create the connection in

play06:05

make.com by creating a web hook and

play06:07

creating our make.com Automation and a

play06:10

lot of the times it'll take 10 or 20

play06:12

minutes to create an automation but it

play06:14

will save you hours of time on the back

play06:16

end and I even give you a GPT to walk

play06:18

you through this step by step and it's

play06:19

right here in a link so it's called the

play06:21

schema ninja and I created this just for

play06:23

this video and what you have the ability

play06:25

to do is hit get request or post request

play06:27

and that will actually send off a set of

play06:30

instructions to the schema ninja in

play06:32

order to help you create that so if we

play06:34

go back to my instructions what you'll

play06:35

notice right now is I need to first set

play06:37

up a get request because right now I say

play06:40

when I say fetch articles get me the

play06:42

most recent articles on AI so how am I

play06:45

going to do that because I need to get

play06:46

that information back into chat GPT

play06:48

therefore it's a get request so I can go

play06:50

to my schema ninja and select get

play06:52

request and it's going to walk you

play06:54

through step byep what you need to do in

play06:56

order to create this request as you can

play06:58

see step one create a hook and make and

play07:00

then change it to these settings right

play07:02

here I even give example images for each

play07:05

step so you can see what I'm talking

play07:07

about as you're using this GPT but keep

play07:09

in mind I'm using a free make account

play07:11

for this okay I'm not using anything

play07:13

paid other than my GPT plus plan which

play07:15

I'm sure if you're watching this you

play07:16

might have right now what we want to do

play07:18

is we want to select create a new

play07:20

scenario and then follow the

play07:21

instructions from the schema ninja so I

play07:23

need to create a web hook and change my

play07:25

settings to get request headers yes and

play07:27

get HTTP method yes so so I can do that

play07:30

by once I create a new scenario this box

play07:32

will pop up I can type in web Hooks and

play07:35

then I want to select custom web hook

play07:38

and here you need to add a web hook so

play07:39

I'm going to hit add you can give it any

play07:41

name you want I'm going to name it

play07:43

posting for

play07:46

LinkedIn beautiful I'll hit show

play07:48

advanced settings and turn on these top

play07:50

two and then hit save then I can hit

play07:53

stop and okay beautiful so now we have

play07:56

the ability to contact this specific URL

play07:59

in order to start an automation that

play08:02

follows this we can start an automation

play08:04

by adding modules here and connecting to

play08:07

so many different apps like Facebook for

play08:09

example or Instagram and we have the

play08:11

ability to download Media get media we

play08:14

could contact notion and create database

play08:16

items we could contact Google Sheets for

play08:20

example and we can create rows we can

play08:22

create sheets create sheets from a

play08:25

template so as you can see there's

play08:26

multiple different things you can do

play08:28

after this web hook URL is contacted

play08:31

with our GPT which I'm going to show you

play08:33

how to set that up now so you can just

play08:35

keep walking through this we've created

play08:37

our web hook in our settings so I'll

play08:38

just say complete and it's going to keep

play08:40

giving you step-by-step instructions for

play08:42

what you need to do so now it says

play08:44

create your automation using make

play08:46

modules so the tool I use for

play08:48

aggregating my news articles is a tool

play08:50

called RSS doapp this allows me to

play08:53

create feeds and connect with

play08:54

information from multiple different

play08:56

sources I have articles pulling in from

play08:58

MIT news artificial intelligence news

play09:01

open AI research and Tech crunch so this

play09:04

URL Aggregates every new post that comes

play09:07

from those websites I can actually

play09:09

connect to that and make so once this

play09:10

web Hook is triggered I can type in RSS

play09:13

and then I can retrieve feed items from

play09:15

a specific feed that I create so I can

play09:17

put in my URL and then I can have a nice

play09:20

custom date from and date to most of you

play09:22

won't be needing this complex of a

play09:24

format for date from and date 2 but I

play09:27

just want to set this up so I can be

play09:28

receiving information from yesterday

play09:31

maximum number of returned items how

play09:32

many articles do I want coming in at

play09:34

once maybe I only want five I can hit

play09:36

okay and now I need to format this data

play09:38

so that it's ready to be put in a web

play09:40

hook response so I have this set up I'm

play09:43

pulling in the URL title and author I

play09:45

can hit okay and now I can aggregate

play09:47

this data in order to pull in each

play09:49

article CU right now it would only pull

play09:51

in one article so I have to make one

play09:53

text strand for all of them so I did

play09:56

that that was very very simple in the

play09:58

text aggregator side of things and many

play09:59

of you won't have to do this complex of

play10:02

tools in order to aggregate some

play10:03

information but next after we've created

play10:05

our automation we can take a look at

play10:08

everything we've done so we've created

play10:10

our web hook and now we need to finish

play10:11

creating our automation so we can do

play10:13

that by just talking to our schema and

play10:15

just some more I'm just going to say

play10:16

complete right there and now the next

play10:18

step after we have our automation done

play10:20

which is these three steps right here

play10:22

what we need to do now is create a web

play10:25

hook response and map our data so the

play10:28

web hook response is important because

play10:29

this is what your GPT is actually going

play10:31

to receive so if I go here and I select

play10:34

web hooks again and I select web hook

play10:37

response in this body field this is what

play10:39

information is going to be given to your

play10:41

GPT so I've set up this information in a

play10:44

way to where all I need to do is get my

play10:47

text from my array aggregator and if I

play10:50

have five articles then I need to make

play10:52

five responses of that array aggregator

play10:55

in the body of that web hook response

play10:58

but you can just put normal text in here

play11:00

too maybe you just want to put hey with

play11:02

a couple exclamation points this

play11:04

response will be received back in your

play11:05

GPT but when you're using these

play11:07

applications you actually have the

play11:09

ability to map specific data to your web

play11:12

hook response and this is where the

play11:14

dynamic ability of your automation comes

play11:16

in play here so I can hit okay beautiful

play11:19

now we have this automation pretty much

play11:22

all up and running we've created our web

play11:24

hook response just as the schema ninja

play11:26

has told us we have an example image

play11:28

here too now I can just say complete

play11:30

what do we do next now we need to create

play11:32

the schema for this get request so we

play11:34

can actually use chat gbt to pull in the

play11:36

information to our GPT here which in

play11:39

this case is the LinkedIn master so if I

play11:42

go back to the checklist actually what

play11:43

we've done is we've created our

play11:44

Automation and now we're on the step

play11:45

three which is creating our GPT action

play11:48

so we need to go back into our GPT that

play11:50

we've created and actually find a way to

play11:52

connect to this automation here so the

play11:55

schema ninja told us to create open AI

play11:57

schema for the get request take a

play11:58

screenshot shot of your web hook

play11:59

response body and upload it here or you

play12:02

can just give it the information it

play12:03

needs in order to create that schema so

play12:07

that it knows which information to pull

play12:09

for now I can just go to my tools here

play12:11

and I can paste in a screenshot of this

play12:13

information that I want chat gbt to be

play12:17

pulling in url title and author so we

play12:20

need to take a screenshot of the

play12:21

information we want pulled in or we can

play12:23

just tell it maybe I want article URL

play12:26

article author article content whatever

play12:28

it may be you can give it all that

play12:30

information that it needs in order to

play12:32

create the schema then you can tell it

play12:33

the goal which one and two kind of go

play12:36

together this is just to get an idea of

play12:37

how to create your schema for the get

play12:39

request and step number three it tells

play12:41

you to provide it with your make web

play12:43

hook URL that you created in the first

play12:46

module right here so this URL this blue

play12:49

URL under your web hook action so I'm

play12:51

just going to keep following this

play12:52

stepbystep process I'm going to upload

play12:55

my screenshot of the things that I

play12:57

wanted pulled in or again you could just

play12:59

tell it the things you want pulled in as

play13:00

well I just want to show you an example

play13:02

of this multimodality and then what you

play13:04

want to do is copy your web hook URL the

play13:06

entire thing and paste it in so I can

play13:09

say I've provided everything create my

play13:14

schema and send it off and it's

play13:16

automatically going to create your

play13:17

schema for you in the Json format that

play13:19

you need this is what you're going to

play13:21

paste in to the action section of your

play13:23

GPT this is that long string of text

play13:26

that is always very intimidating

play13:27

whenever you want to automate something

play13:28

so we we go back to our GPT now and we

play13:31

scroll all the way down to the bottom

play13:33

where it says actions down here we have

play13:36

the ability to hit create new action

play13:39

here we need to upload the schema that

play13:40

was just created by the schema ninja

play13:43

based on all the properties you want

play13:45

pulled in from your automation so I can

play13:47

copy this code head back to chat gbt and

play13:50

paste in the schema just like that we

play13:52

now have a get request with our specific

play13:55

path our operation ID and it's very

play13:58

beautiful so now we can even test this

play14:01

but before we test it we can just go

play14:03

back to our schema ninja and we can say

play14:06

complete now that we've uploaded this

play14:08

there was no errors nothing went wrong

play14:10

we have a beautiful action in here ready

play14:12

to go and this right here when we hit

play14:14

test is going to contact this web hook

play14:17

which will trigger the automation we

play14:19

just created and after it runs through

play14:21

this three-step process in the middle of

play14:23

the red modules what it will do is send

play14:25

back what it got from running through

play14:27

that process okay so I'm going to turn

play14:29

this on in the bottom left hand corner

play14:32

and then I'm going to hit save on my

play14:33

automation because once you hit complete

play14:35

in your schema ninja that's what it's

play14:37

going to tell you to do is to save and

play14:39

test your information then we can send

play14:41

off a test so I'm going to hit test and

play14:43

what it should do is pull in all the

play14:46

information that we're getting from my

play14:48

RSS feed which is an external app that I

play14:50

wanted connected so I'm going to hit

play14:52

always allow up here and it's going to

play14:54

talk to this web hook it's giving me the

play14:56

post now from my RSS feed Fe so this

play14:59

information is very custom because it's

play15:01

coming from my specific feed that I

play15:03

wanted in that external app and it's

play15:05

actually making LinkedIn post out of

play15:07

these right now so the data came in but

play15:09

maybe it didn't come in in the format

play15:10

you want or maybe it didn't come in in

play15:12

the exact way you wanted as long as you

play15:15

get response received with a status code

play15:17

of 200 then it means your automation is

play15:19

working you have the ability to change

play15:22

whatever comes after this web hook

play15:24

update your schema that you've created

play15:27

with the schema ninja and actually pull

play15:29

in whatever information you'd like based

play15:31

on this web hook URL as you can see

play15:33

though it's just giving me the post and

play15:35

it's not showing me the title author and

play15:37

URL first maybe first I want to actually

play15:40

read the article then tell it to create

play15:42

a post based around it so I can change

play15:44

the instructions and edited around at

play15:46

this point but we've completed our get

play15:48

request so as you can see I've added in

play15:49

some very basic instruction updates I

play15:51

say before you create LinkedIn post

play15:53

provide me with the following author

play15:55

author here title Title Here url url

play15:58

here then ask me which article would you

play16:00

like to create a LinkedIn post around

play16:01

that way I have a little bit more

play16:02

control and it's a little bit more of a

play16:04

process rather than just creating

play16:06

articles and post right from the jump so

play16:09

now I can go back and I can test this

play16:11

again by hitting fetch articles now that

play16:13

we have our get request set up since we

play16:15

set up the instructions to contact the

play16:17

get request once we say fetch articles

play16:19

our conversation starters will be

play16:20

working so I can hit fetch articles and

play16:22

test this out again to see if it's

play16:23

pulling in data how I want it to so as

play16:25

you can see now it's coming back with

play16:27

everything I asked for it's giving me

play16:28

the author and then the title and the

play16:31

title is actually in the form of a

play16:32

hyperlink which is beautiful so now I

play16:33

can actually click on these articles and

play16:36

go to the source where they're coming

play16:38

from and then it ask me which article

play16:39

would you like to create a LinkedIn post

play16:41

around I could just type out a number at

play16:43

this point I could just say number three

play16:44

for instance then it's actually going to

play16:46

create that post around the article so

play16:48

it's starting off with what if we could

play16:50

design proteins to revolutionize

play16:52

medicine and Beyond going a little bit

play16:54

in depth and then having a CTA and then

play16:56

it asks me the question would you like

play16:58

me to send this as your LinkedIn post I

play17:00

could say yes but we don't have that

play17:02

action set up yet so if we go back to

play17:03

our checklist we now have the ability to

play17:05

check off the remaining things because

play17:07

we've created our GPT action and we've

play17:09

tested our GPT action we've grabbed test

play17:11

data from the get request now what if we

play17:13

actually want to send these LinkedIn

play17:15

post well then we need to revise our

play17:17

schema because our schema right now if

play17:19

we go to our actions only has one method

play17:24

of using the HTTP request which is get

play17:26

right now so we only have one avail

play17:29

action right here and that's to get web

play17:31

hook data or in other words get our

play17:33

feeds into our news articles from that

play17:36

app what if we wanted to send this well

play17:38

then we would need to add to our

play17:40

Automation and we'd also need to revise

play17:44

that schema in order to make it

play17:46

available to have another method of HTTP

play17:49

request and in order to edit this first

play17:51

we need to have test data with our web

play17:53

hook so we can dynamically map fields

play17:56

and the web hook data needs to be the

play17:58

LinkedIn post we need to figure out a

play18:00

way to pull in Dynamic LinkedIn posts to

play18:02

this web hook so we can do that very

play18:04

simply first we can go to our schema

play18:08

ninja GPT and type in a new prompt and

play18:11

make sure you're staying in the same

play18:12

chat thread for this so what I'm doing

play18:14

here is I'm basically staying in the

play18:17

same thread because we know that this

play18:19

schema is working and this GPT is

play18:22

designed to create schema that works so

play18:24

what I can do very simply is add in

play18:27

another http method to the schema that's

play18:30

already working and that and it already

play18:32

exists and then later we can change

play18:34

around our automation after we send it

play18:35

some test data but I say using this same

play18:38

schema add in a poster request with the

play18:41

get request I want this poster request

play18:42

to be able to send my content to the web

play18:44

hook the content in this case would be a

play18:46

LinkedIn post help me revise the web

play18:48

hook in order to be able to add another

play18:50

HTTP method of post where I can send

play18:53

information to my web hook in the form

play18:55

of a post I keep reiterating in the form

play18:57

of a post so at this step if you want to

play18:59

not only get information into your GPT

play19:01

but be able to post it elsewhere which

play19:03

some people are fine with just pulling

play19:04

in the information and then working with

play19:06

it and then doing whatever with it later

play19:08

a lot of the times that's the hardest

play19:09

part is getting that information in but

play19:11

we can actually revise our schema to add

play19:13

a post action to do something else so I

play19:16

can send this off I recommend just using

play19:18

a similar prompt if you're trying to add

play19:20

to your schema to get a post request

play19:22

query but what it's going to do is it's

play19:24

going to use the same URL it's going to

play19:26

use your same web hook path except

play19:28

underneath get it will have your entire

play19:31

get operation and then it's going to go

play19:33

into a post operation so we're going to

play19:35

send a LinkedIn post to our web hook

play19:38

very very cool it gives us our updated

play19:40

schema so right away what we can do is

play19:42

we can copy this updated schema that it

play19:44

gives us go to our GPT and actually

play19:47

paste it in but first you want to delete

play19:49

your old schema which might be a little

play19:51

scary but then you can paste in new and

play19:54

what's going to happen is you're going

play19:55

to see another method down here so now

play19:57

we have the ability with our GPT to get

play19:59

information and post information but as

play20:01

you can see we have some required Fields

play20:03

down here it says post title and post

play20:05

content this is what we want our GPT to

play20:07

format our information with we want a

play20:10

post title and post content in order to

play20:12

actually send these posts to our web

play20:15

hook so now we need to go edit our

play20:17

instructions to ensure that whenever it

play20:19

creates a LinkedIn post it formats it in

play20:21

this Json format and it actually with

play20:23

our schema ninja gives us that format

play20:25

that we need and it gives us

play20:27

instructions for a poster request so

play20:29

what I can do is I can just copy these

play20:31

instructions head back to our GPT and

play20:34

then go paste those instructions in our

play20:37

instructions so I can leave these at the

play20:39

very bottom and these are going to be

play20:41

instructions for the Post request when

play20:43

using this post request you'll need to

play20:44

send a Json body that contains post

play20:47

title I can put those in quotations and

play20:50

post content Fields because you need a

play20:52

structured way to send your data to your

play20:54

web hook in order to be able to use that

play20:56

data dynamically so that's why we're

play20:58

doing it in Json format so underneath

play21:00

these instructions for post request I

play21:02

can then put that example Json body I

play21:04

can say here is an example of how to

play21:08

structure this data and I can go back to

play21:11

the schema ninja and copy this code the

play21:14

example Json body for post request and

play21:15

if it didn't give this to you just ask

play21:17

for this stuff and the schema ninja is

play21:19

really good at doing that for you and

play21:21

then I can paste it in beautiful we have

play21:23

the structure of how this should look I

play21:25

can then hit close and now we can go

play21:27

test our post request to make sure it's

play21:29

actually running so I can go to actions

play21:31

and under available actions where it

play21:33

says post I want to test this so I can

play21:36

hit test on post and it will send an

play21:39

example post to your automation on make

play21:42

just so we have data to pull from when

play21:44

we do create the rest of this automation

play21:46

I can hit confirm and I can kind of run

play21:49

through this process I'll just say

play21:50

article three and then I can say post

play21:53

and when I send that off it should

play21:54

perform the action of the post method

play21:57

and it says the LinkedIn post has

play21:58

successfully been sent what I'm going to

play22:00

do now is actually map this data to our

play22:02

web hook and create a LinkedIn post

play22:04

around it like physically create the

play22:06

automation that we can use from the data

play22:09

that we're receiving from our GPT here

play22:12

so I'm going to unlink in between right

play22:15

here and I want to have a router because

play22:18

a router is going to allow us to go to

play22:21

different methods so I can select router

play22:23

and this is going to be the router to go

play22:25

to our get request this will be a

play22:26

fallback method and I'll explain what

play22:28

that that means in a minute but now I

play22:29

can add the Linked In and once I hit

play22:33

LinkedIn I can just do create a user

play22:35

text post and add that to the router so

play22:37

now we have two different routes to go

play22:39

the post request method and the get

play22:40

request method in order to make sure

play22:42

that it's filtering out correctly what

play22:45

we need to do is set up a filter on our

play22:47

post request we can select the wrench

play22:50

hit set up a filter label this can just

play22:52

be condition setting and then for

play22:56

condition we can have meth

play22:58

is equal to post and we can just make

play23:02

equal to case insensitive and then I can

play23:04

hit okay so now if it is a post it will

play23:08

go here first and this will be a

play23:10

fallback route we can set up this filter

play23:12

in between the RSS feed and the router

play23:14

to be a fallback route and this will be

play23:17

the first condition it sees and checks

play23:19

if it can go to next what you need to do

play23:21

is you just need to hit send data again

play23:23

this is just going to make sure that

play23:24

your data is being sent in your test

play23:26

feed and I just do this in order to make

play23:28

sure the fields are pulling in because

play23:30

sometimes when you go here and your

play23:32

automation isn't saved when you try to

play23:34

map in your content to the dynamic

play23:36

fields in your web hook they won't pull

play23:38

in so what you need to do if it's not

play23:40

pulling in is save your Automation and

play23:42

hit this little back arrow and then go

play23:44

to your GPT where you've been testing

play23:46

and just hit send data again and then

play23:48

when you click back in here you should

play23:50

be able to dynamically pull in that

play23:52

content so now I can pull in post

play23:54

content just like that so this would be

play23:56

an example of setting up a post request

play23:58

we need to be able to pull in

play24:00

information dynamically from our web

play24:02

hook that's coming from our GPT like

play24:03

shown I can hit okay I can save this

play24:06

Automation and now we can put this to

play24:07

the test I'm going to go to my checklist

play24:10

check off the post request again you

play24:12

don't have to do a post request optional

play24:14

depending on your Automation and then

play24:15

what I can do is I can come to my GPT

play24:17

now that all the testing is done in the

play24:18

upper right hand corner I can hit create

play24:21

and then I can make access invite only

play24:24

and then I can hit update or I can

play24:25

create it for the first time if that's

play24:27

what you're doing then I can hit view

play24:29

GPT next I can just walk through this

play24:31

process and do it so I can hit fetch

play24:33

articles then I can hit always allow on

play24:35

that web hook because it needs

play24:36

permission to talk to it and that's

play24:38

going to give me all the recent articles

play24:40

from my feed that I can create post

play24:42

around based on the last 254 hours so

play24:45

maybe I could say post number three and

play24:48

send that off and now we have this

play24:50

automated process to where I don't even

play24:51

have to leave my GPT screen and it's

play24:53

going to give me a post draft it's going

play24:55

to give me this beautiful LinkedIn post

play24:57

based on the article and I could say

play25:00

revise it to attribute content to the

play25:04

author at the end of the post you can

play25:07

even revise these posts and you can

play25:09

revise your content and chaty BT is

play25:11

basically your assistant now you can ask

play25:13

it to go get things for you then you can

play25:14

ask it to post things for you so I'm

play25:16

telling you to provide a link of the

play25:17

article at the bottom instead of the

play25:19

inspired by comment because I don't want

play25:20

it to just say content inspired by I

play25:22

want it to provide a link to where I am

play25:24

uh getting these thoughts or these ideas

play25:26

from so now it ask me would you like to

play25:28

send this post to the web hook for

play25:29

LinkedIn posting I can say yes

play25:32

absolutely so as you can see I'll just

play25:33

have to hit confirm really quick very

play25:35

easy and then it's going to post that on

play25:37

LinkedIn your LinkedIn post has been

play25:39

successfully published it used the

play25:41

information that it got from here very

play25:43

quick and a very automated system as you

play25:45

can see if I go on my LinkedIn and go to

play25:48

posted just now I can click this and it

play25:50

did post literally everything in a

play25:53

matter of seconds just like that that's

play25:55

how Innovative this is that's how

play25:57

efficiency works within AI now if you

play26:00

enjoyed this and you want to even become

play26:02

more efficient then again I recommend

play26:04

the AI foundations Community to anybody

play26:07

I mean since just recording this video

play26:09

we already have two new members in the

play26:10

past hour people are joining in order to

play26:12

leverage efficiency with AI and people

play26:14

that are applying this stuff in their

play26:16

business and their personal life are

play26:17

succeeding with it there's so many

play26:19

different Industries in here leveraging

play26:20

artificial intelligence and you could be

play26:22

one of them so if you want to learn more

play26:24

cool things like this to save you hours

play26:25

of your time using this technology of

play26:27

the future then I don't want you to fall

play26:29

behind and I want you to join the

play26:30

community we have so much to offer there

play26:32

live calls courses Etc in order to get

play26:35

you from A to B but with that being said

play26:36

I hope you enjoyed this Automation and I

play26:38

hope it helped you if it did Please

play26:39

Subscribe and like this video I will be

play26:41

coming out with more content like this

play26:43

if you'd like me to go more in depth

play26:44

just let me know I be happy to answer

play26:47

your comments record more videos

play26:49

automate more things so I can help save

play26:50

you time with AI all right I'll see you

play26:53

in the next video

Rate This

5.0 / 5 (0 votes)

Связанные теги
AI-AssistentenAutomatisierungEffizienzProduktivitätGPTChatbotZeit sparenDigitalisierungAI-TechnologieWorkflow
Вам нужно краткое изложение на английском?