Automated YouTube Video Scheduling & AI Metadata Generation 🎬

Build Challenge

Drag nodes onto the canvas, connect them in order, then grade or reveal the solution.

n8n

Make

Zapier

Activepieces

Power Automate

🔴 n8n Workflow

Drag-and-drop node canvas

⚡

Webhook

Trigger

✏️

Set Fields

Build prompt

🤖

OpenAI

generateText

📤

Respond

Return output

🏗️ Guided Build — n8n

Step-by-step

Add a Webhook trigger node

Create a new workflow, add a Webhook node. Set the HTTP Method to POST and copy the test URL. Click "Listen for test event" before sending your first request.

💡 Use the Production URL once the workflow is active — the test URL only works while you're listening.

Build the prompt with a Set node

Add a Set node to assemble your prompt string. Use expressions like {{ $json.topic }} to inject incoming data. Store the result in a field called prompt.

💡 Keep system and user prompts as separate fields — it's cleaner to map them individually into the OpenAI node.

Connect the OpenAI node

Add an OpenAI node, select the "Message a Model" operation. Map your {{ $json.prompt }} field into the User Message. Set the model to gpt-4o and Max Tokens to 512.

💡 Add a system prompt field to the Set node and map it to the System Message — this keeps your instructions separate from user input.

Return the result with Respond to Webhook

Add a "Respond to Webhook" node. Set Response Body to {{ $json.message.content }} from the OpenAI output. Activate the workflow and test end-to-end.

💡 Set Response Code to 200 and Content-Type to application/json — callers expect a proper response, not an empty 200.

🎯 Sandbox Mission

Extend it in n8n

Add a tone parameter

Modify the workflow so the caller can pass a tone field (e.g. "professional", "casual", "witty"). Inject it into the system prompt so the model adapts its output accordingly.

Add an IF node after Set — if tone is empty, default to "professional". Otherwise use the caller's value.

🟠 Make Scenario

Mapped from n8n

⚡

Webhooks

Custom webhook

✏️

Tools

Set Variable

🤖

OpenAI

Create Completion

📤

Webhooks

Send response

🏗️ Guided Build — Make

Step-by-step

Create a scenario with a Custom Webhook

Click Create a new scenario. Search for Webhooks and add Custom webhook as the first module. Save to generate the URL, then click "Run once" to start listening for a test payload.

💡 POST a JSON body with a prompt key to capture the data structure for mapping in later modules.

Set Variable to build your prompt string

Add a Tools → Set Variable module. Create a variable named fullPrompt and map the webhook's prompt field into it. You can concatenate a system instruction here too.

💡 Make doesn't have a native "system prompt" field in all OpenAI modules — prepend your instruction directly: "You are X. User says: {{prompt}}"

Add OpenAI → Create a Completion

Search for OpenAI and add "Create a Completion". Set Model to gpt-4o, map your fullPrompt variable into the Prompt field. Set Max Tokens to 512 and Temperature to 0.7.

💡 Always set Max Tokens explicitly in Make — leaving it blank can result in truncated or empty outputs with no error shown.

Return the result via Webhooks → Send response

Add a Webhooks → Send response module. Set Status to 200, Body Type to JSON, and paste {"result": "{{choices[].text}}"} mapping the OpenAI output text. Toggle the scenario active.

💡 Run the scenario once manually before activating to confirm the full data flow end to end.

🎯 Sandbox Mission

Extend it on Make

Add a Router for multiple output formats

After the OpenAI module, add a Router. Route 1 returns the raw text via webhook. Route 2 appends the result to a Google Sheet row. Wire both routes and verify both paths execute cleanly.

Right-click the OpenAI module → Add route. Each route is independent — the Router fans output to all active routes simultaneously.

⚡ Zapier Zap

Mapped from n8n

⚡

Trigger

Webhooks by Zapier

Catch Hook

TRIGGER

✏️

Action

Formatter by Zapier

Text — Truncate

🤖

Action

OpenAI (GPT-4, DALL-E, Whisper)

Send Prompt

📤

Action

Webhooks by Zapier

Return Response

⬇ Export Zap (requires Zapier account)

🏗️ Guided Build — Zapier

Step-by-step

Create a Zap with Webhooks trigger

Click Create Zap. Choose Webhooks by Zapier as the trigger, select Catch Hook. Click Continue and copy the webhook URL. POST a test JSON body containing a prompt key to it.

💡 Webhooks by Zapier requires a paid plan. On free, use Google Sheets: New Row as trigger and put your prompt in a sheet column instead.

Clean the prompt with Formatter

Add Formatter by Zapier → Text → Truncate. Set the input to the prompt field from step 1, Max Length to 2000. This prevents oversized inputs from hitting the OpenAI token limit.

💡 Formatter is also useful here for "Remove HTML" if the prompt could contain markup from a web form.

Add OpenAI action — Send Prompt

Search for OpenAI (GPT-4, DALL-E, Whisper) and choose Send Prompt. Map the truncated text into the Prompt field. Choose gpt-4o, set Max Tokens to 512 and Temperature to 0.7.

💡 The "User Name" field in this Zapier action is optional — but populating it helps with audit logs if multiple users share the Zap.

Return the result via Webhooks Response

Add Webhooks by Zapier → Return Response. Set Status Code to 200 and paste the OpenAI response text into the body field. Publish the Zap and test the full flow.

💡 Return Response only works when the Zap was triggered by a Webhook Catch Hook — other triggers won't have a return path.

🎯 Sandbox Mission

Extend it on Zapier

Add a Filter step for short prompts

Insert a Zapier Filter between Formatter and OpenAI. Only continue if the prompt length is greater than 10 characters. If it's too short, the Zap stops — no wasted OpenAI tokens on empty inputs.

Add Filter by Zapier → Only continue if → Text Length → Greater than → 10.

🧩 Activepieces Flow

Mapped from n8n

⚡

HTTP

Webhook Trigger

⚙️

Code

Build prompt

🤖

OpenAI

Ask ChatGPT

📤

HTTP

Return response

⬇ Export Flow (from Activepieces dashboard)

🏗️ Guided Build — Activepieces

Step-by-step

Add the Webhook trigger piece

Create a new flow. Click Add Trigger, search for HTTP and select Webhook. Copy the generated URL. The webhook URL is stable — it won't change between test runs.

💡 POST a body like {"prompt":"Write a haiku about automation"} to capture the shape for mapping downstream.

Use a Code piece to build the prompt

Add a Code step. Write a small JS function to assemble a messages array: [{role:"system",content:"..."},{role:"user",content:trigger.body.prompt}]. Return it as messages.

💡 Activepieces Code steps run server-side Node.js. You can use fetch directly if you need a custom API call instead of a built-in piece.

Add OpenAI — Ask ChatGPT

Add the OpenAI piece and select "Ask ChatGPT". Map your Code step's messages output into the Messages field. Set model to gpt-4o and Max Tokens to 512.

💡 Activepieces keeps the same API connection across all OpenAI pieces — connect credentials once in Settings → Connections.

Return the response to the caller

Add an HTTP piece → Return Response. Set status 200, body type JSON, and map {{openai.choices[0].message.content}}. Publish the flow and run a live test.

💡 Activepieces shows a live Run Log after each test — expand each step to see input and output data at that point in the flow.

🎯 Sandbox Mission

Extend it on Activepieces

Log every request to a NocoDB table

After the OpenAI step, add a NocoDB piece to insert a row: store the incoming prompt, the model's response, and a timestamp. This gives you a persistent audit trail of every generate_text call.

Use Activepieces' NocoDB piece → Create Row. Map fields: prompt, result, createdAt (use {{now}} for timestamp).

🔵 Power Automate Flow

Mapped from n8n

⚡

Trigger

When a HTTP request is received

Instant cloud flow

TRIGGER

✏️

Action

Initialize variable

String — promptText

🤖

Action

HTTP — POST to OpenAI

api.openai.com/v1/chat/completions

📤

Action

Response

Return 200 + body

⬇ Export Package (from Power Automate → Export)

🏗️ Guided Build — Power Automate

Step-by-step

Create an Instant cloud flow with HTTP trigger

In Power Automate click Create → Instant cloud flow. Search for "HTTP request" and select "When a HTTP request is received". Click Save — the POST URL only appears after the first save.

💡 Paste a sample JSON schema like {"prompt":"string"} into the schema field — Power Automate auto-parses it into named variables.

Initialize a variable for the prompt

Add the "Initialize variable" action. Name it promptText, type String, and set the value to triggerBody()?['prompt']. This extracts the prompt from the incoming JSON cleanly.

💡 Always initialize variables near the top — Power Automate expressions can't reference a variable before it's declared in the flow order.

Call the OpenAI API via HTTP action

Add an HTTP action. Method: POST, URI: https://api.openai.com/v1/chat/completions. Add header Authorization: Bearer YOUR_KEY. Body: {"model":"gpt-4o","messages":[{"role":"user","content":"@{variables('promptText')}"]}.

💡 Store the API key in an Environment Variable or Azure Key Vault reference — never hardcode it in the flow body.

Return the result with a Response action

Add a Response action. Status code 200. Body: @{body('HTTP')?['choices']?[0]?['message']?['content']}. This extracts the text from the OpenAI JSON response and returns it to the caller.

💡 Wrap the HTTP and Response in a Scope named "OpenAI Call" — makes the flow readable and gives you a clean place to add error handling later.

🎯 Sandbox Mission

Extend it on Power Automate

Add error handling with a Scope

Wrap the HTTP action inside a Scope. Add a second Scope set to run "if previous step fails". Inside it, add a Response action that returns status 500 and a JSON error body. This makes your flow resilient to OpenAI outages or bad keys.

Configure the error Scope: click the three dots → Configure run after → check "has failed". This is Power Automate's equivalent of a try/catch.

🧠 LLM Compatibility

Recommended

GPT-4o / GPT-4o-mini

OpenAI

FastLow costJSON mode

Best balance of speed/quality for email drafting and summarisation.

Claude 3.5 Sonnet

Anthropic

Long contextSafe

Excellent for long documents, nuanced tone, and multi-turn support.

Gemini 1.5 Pro

Google

1M contextMultimodal

Use when processing large documents or needing image + text in one step.

Mistral 7B / Mixtral

Mistral AI

Open weightSelf-host

Run locally via Ollama for zero API cost. Best for simple classification and drafts.

Llama 3.1 70B

Meta via Ollama

FreeLocal

Strong open model for production quality replies. Requires GPU server.

All models work with the n8n OpenAI, Anthropic, Google Gemini, and Ollama nodes. Swap models without changing workflow logic — only credentials and model name change.

🏆 Challenge: Automated YouTube Video Scheduling & AI Metadata Generation 🎬 — Automation University

Recreate the core node chain for this workflow

Study the animated playback above, then build the key node sequence in the canvas below. Match the trigger and processing steps this workflow uses.

⚡ Trigger ⚙️ Process 📤 Output Connect them left → right

Triggers

🔗

Webhook

Trigger

📧

Read Email

Trigger

✍️

Generate Text

OpenAI / Claude

🏷️

Classify

AI Label

Output

📬

Send Email

Gmail / SMTP

📨

Send Message

Slack / Telegram

📤

Return Output

Response

Data

🔍

Extract Data

Parse fields

⚙️

Transform

Set / Code

↙️

Drag nodes from the left panel

Click another node to connect →

📊 Live Stats

~2h 34min

Est. Build Time

6.5/10

Difficulty

+1555 XP

On Completion

Nodes

Builds Completed

Rebuilt from Scratch

Practice Passes

Capstones Shipped

Loading live stats...

⚠️ Common Mistakes

From workflow analysis

❶

System prompt specificity: Vague system prompts produce inconsistent results. Be explicit about output format, tone, and what the AI should NOT do.

❷

Memory node port: Connect memory via the ai_memory port on the Agent node — NOT the main execution port. Wrong port = no memory.

❸

Token overflow: Large inputs can exceed context limits. Add a Code node to trim or summarise data before passing to the AI model.

❹

Temperature for structured output: Use temperature 0 when the AI needs to produce structured or deterministic output. Higher values introduce unpredictable formatting.

🗺️ Active Quests

Complete objectives · Earn XP · Track mastery

🏆 Achievements

🏗️

Build Complete

Finish the full guided build to earn this.

🧩

Mock Builder

Rebuild from scratch and submit to earn this.

🎯

Practice Pro

Score 70%+ on practice questions.

👑

Capstone Master

Ship the workflow in production.

4,038

Workflows using this skill

57%

of all AI workflows

compatible node types

most common AI skill

What is Generate Text?

🧠

Core Concept

Generate Text means invoking a language model (GPT-4, Claude, Gemini, Mistral) inside an n8n workflow to produce a natural language output — a summary, reply, classification label, code snippet, or structured JSON from a prompt.

🎯

When to Use

Use this skill when the output of a step needs to be written language: email drafts, support replies, content summaries, translation, sentiment labels, entity extraction, or any task where a human would normally read and write.

⚡

Key Inputs

Every Generate Text step needs a prompt (the instruction) and optionally context (documents, previous messages, user data). The prompt is assembled from upstream node output using n8n expressions like {{ "{{" }}$json.body{{ "}}" }}.

📤

Key Output

Returns a text string in $json.text or $json.content depending on the node. You can chain the output into Send Email, Write Record, Send Message, or any node that accepts text.

Common Patterns

Webhook → Generate Text → Send Email

Receive a form submission, generate a personalised reply with GPT-4, send it back. Powers auto-responders, lead follow-up, and support triaging.

Read Email → Generate Text → Write Record

Parse inbox messages, extract entities or classify intent, log structured output to a database. Reduces manual data entry to zero.

Read File → Generate Text → Return Output

Ingest PDFs or CSVs, summarise contents with an LLM, return the digest to another system or display in a dashboard.

Receive Webhook → Memory Retrieval → Generate Text → Reply

The conversational agent pattern. Retrieves prior context from a vector store, augments the prompt, generates a context-aware reply. Core of every AI chatbot built in n8n.

Loop Items → Generate Text → Aggregate Data

Process a list (products, tickets, articles) through an LLM in batch. Generate a summary or label for each item, then aggregate results into one output.

Compatible Nodes

🤖

OpenAI

GPT-4, GPT-4o

🧠

Anthropic

Claude 3.5+

💎

Google Gemini

Gemini Pro

🌐

HTTP Request

Custom LLM API

⛓️

LangChain Agent

Tool-calling

🔗

Ollama

Local LLMs

🪄

Mistral

Mistral 7B+

🔮

Cohere

Command R

Build Challenge

Drag nodes onto the canvas, connect them in order, then grade or reveal the solution.

n8n

Make

Zapier

Activepieces

Power Automate

🔴 n8n Workflow

Drag-and-drop node canvas

⚡

Webhook

Trigger

✏️

Set Fields

Build prompt

🤖

OpenAI

generateText

📤

Respond

Return output

🏗️ Guided Build — n8n

Step-by-step

Add a Webhook trigger node

Create a new workflow, add a Webhook node. Set the HTTP Method to POST and copy the test URL. Click "Listen for test event" before sending your first request.

💡 Use the Production URL once the workflow is active — the test URL only works while you're listening.

Build the prompt with a Set node

Add a Set node to assemble your prompt string. Use expressions like {{ $json.topic }} to inject incoming data. Store the result in a field called prompt.

💡 Keep system and user prompts as separate fields — it's cleaner to map them individually into the OpenAI node.

Connect the OpenAI node

Add an OpenAI node, select the "Message a Model" operation. Map your {{ $json.prompt }} field into the User Message. Set the model to gpt-4o and Max Tokens to 512.

💡 Add a system prompt field to the Set node and map it to the System Message — this keeps your instructions separate from user input.

Return the result with Respond to Webhook

Add a "Respond to Webhook" node. Set Response Body to {{ $json.message.content }} from the OpenAI output. Activate the workflow and test end-to-end.

💡 Set Response Code to 200 and Content-Type to application/json — callers expect a proper response, not an empty 200.

🎯 Sandbox Mission

Extend it in n8n

Add a tone parameter

Modify the workflow so the caller can pass a tone field (e.g. "professional", "casual", "witty"). Inject it into the system prompt so the model adapts its output accordingly.

Add an IF node after Set — if tone is empty, default to "professional". Otherwise use the caller's value.

🟠 Make Scenario

Mapped from n8n

⚡

Webhooks

Custom webhook

✏️

Tools

Set Variable

🤖

OpenAI

Create Completion

📤

Webhooks

Send response

🏗️ Guided Build — Make

Step-by-step

Create a scenario with a Custom Webhook

Click Create a new scenario. Search for Webhooks and add Custom webhook as the first module. Save to generate the URL, then click "Run once" to start listening for a test payload.

💡 POST a JSON body with a prompt key to capture the data structure for mapping in later modules.

Set Variable to build your prompt string

Add a Tools → Set Variable module. Create a variable named fullPrompt and map the webhook's prompt field into it. You can concatenate a system instruction here too.

💡 Make doesn't have a native "system prompt" field in all OpenAI modules — prepend your instruction directly: "You are X. User says: {{prompt}}"

Add OpenAI → Create a Completion

Search for OpenAI and add "Create a Completion". Set Model to gpt-4o, map your fullPrompt variable into the Prompt field. Set Max Tokens to 512 and Temperature to 0.7.

💡 Always set Max Tokens explicitly in Make — leaving it blank can result in truncated or empty outputs with no error shown.

Return the result via Webhooks → Send response

Add a Webhooks → Send response module. Set Status to 200, Body Type to JSON, and paste {"result": "{{choices[].text}}"} mapping the OpenAI output text. Toggle the scenario active.

💡 Run the scenario once manually before activating to confirm the full data flow end to end.

🎯 Sandbox Mission

Extend it on Make

Add a Router for multiple output formats

After the OpenAI module, add a Router. Route 1 returns the raw text via webhook. Route 2 appends the result to a Google Sheet row. Wire both routes and verify both paths execute cleanly.

Right-click the OpenAI module → Add route. Each route is independent — the Router fans output to all active routes simultaneously.

⚡ Zapier Zap

Mapped from n8n

⚡

Trigger

Webhooks by Zapier

Catch Hook

TRIGGER

✏️

Action

Formatter by Zapier

Text — Truncate

🤖

Action

OpenAI (GPT-4, DALL-E, Whisper)

Send Prompt

📤

Action

Webhooks by Zapier

Return Response

⬇ Export Zap (requires Zapier account)

🏗️ Guided Build — Zapier

Step-by-step

Create a Zap with Webhooks trigger

Click Create Zap. Choose Webhooks by Zapier as the trigger, select Catch Hook. Click Continue and copy the webhook URL. POST a test JSON body containing a prompt key to it.

💡 Webhooks by Zapier requires a paid plan. On free, use Google Sheets: New Row as trigger and put your prompt in a sheet column instead.

Clean the prompt with Formatter

Add Formatter by Zapier → Text → Truncate. Set the input to the prompt field from step 1, Max Length to 2000. This prevents oversized inputs from hitting the OpenAI token limit.

💡 Formatter is also useful here for "Remove HTML" if the prompt could contain markup from a web form.

Add OpenAI action — Send Prompt

Search for OpenAI (GPT-4, DALL-E, Whisper) and choose Send Prompt. Map the truncated text into the Prompt field. Choose gpt-4o, set Max Tokens to 512 and Temperature to 0.7.

💡 The "User Name" field in this Zapier action is optional — but populating it helps with audit logs if multiple users share the Zap.

Return the result via Webhooks Response

Add Webhooks by Zapier → Return Response. Set Status Code to 200 and paste the OpenAI response text into the body field. Publish the Zap and test the full flow.

💡 Return Response only works when the Zap was triggered by a Webhook Catch Hook — other triggers won't have a return path.

🎯 Sandbox Mission

Extend it on Zapier

Add a Filter step for short prompts

Insert a Zapier Filter between Formatter and OpenAI. Only continue if the prompt length is greater than 10 characters. If it's too short, the Zap stops — no wasted OpenAI tokens on empty inputs.

Add Filter by Zapier → Only continue if → Text Length → Greater than → 10.

🧩 Activepieces Flow

Mapped from n8n

⚡

HTTP

Webhook Trigger

⚙️

Code

Build prompt

🤖

OpenAI

Ask ChatGPT

📤

HTTP

Return response

⬇ Export Flow (from Activepieces dashboard)

🏗️ Guided Build — Activepieces

Step-by-step

Add the Webhook trigger piece

Create a new flow. Click Add Trigger, search for HTTP and select Webhook. Copy the generated URL. The webhook URL is stable — it won't change between test runs.

💡 POST a body like {"prompt":"Write a haiku about automation"} to capture the shape for mapping downstream.

Use a Code piece to build the prompt

Add a Code step. Write a small JS function to assemble a messages array: [{role:"system",content:"..."},{role:"user",content:trigger.body.prompt}]. Return it as messages.

💡 Activepieces Code steps run server-side Node.js. You can use fetch directly if you need a custom API call instead of a built-in piece.

Add OpenAI — Ask ChatGPT

Add the OpenAI piece and select "Ask ChatGPT". Map your Code step's messages output into the Messages field. Set model to gpt-4o and Max Tokens to 512.

💡 Activepieces keeps the same API connection across all OpenAI pieces — connect credentials once in Settings → Connections.

Return the response to the caller

Add an HTTP piece → Return Response. Set status 200, body type JSON, and map {{openai.choices[0].message.content}}. Publish the flow and run a live test.

💡 Activepieces shows a live Run Log after each test — expand each step to see input and output data at that point in the flow.

🎯 Sandbox Mission

Extend it on Activepieces

Log every request to a NocoDB table

After the OpenAI step, add a NocoDB piece to insert a row: store the incoming prompt, the model's response, and a timestamp. This gives you a persistent audit trail of every generate_text call.

Use Activepieces' NocoDB piece → Create Row. Map fields: prompt, result, createdAt (use {{now}} for timestamp).

🔵 Power Automate Flow

Mapped from n8n

⚡

Trigger

When a HTTP request is received

Instant cloud flow

TRIGGER

✏️

Action

Initialize variable

String — promptText

🤖

Action

HTTP — POST to OpenAI

api.openai.com/v1/chat/completions

📤

Action

Response

Return 200 + body

⬇ Export Package (from Power Automate → Export)

🏗️ Guided Build — Power Automate

Step-by-step

Create an Instant cloud flow with HTTP trigger

In Power Automate click Create → Instant cloud flow. Search for "HTTP request" and select "When a HTTP request is received". Click Save — the POST URL only appears after the first save.

💡 Paste a sample JSON schema like {"prompt":"string"} into the schema field — Power Automate auto-parses it into named variables.

Initialize a variable for the prompt

Add the "Initialize variable" action. Name it promptText, type String, and set the value to triggerBody()?['prompt']. This extracts the prompt from the incoming JSON cleanly.

💡 Always initialize variables near the top — Power Automate expressions can't reference a variable before it's declared in the flow order.

Call the OpenAI API via HTTP action

💡 Store the API key in an Environment Variable or Azure Key Vault reference — never hardcode it in the flow body.

Return the result with a Response action

Add a Response action. Status code 200. Body: @{body('HTTP')?['choices']?[0]?['message']?['content']}. This extracts the text from the OpenAI JSON response and returns it to the caller.

💡 Wrap the HTTP and Response in a Scope named "OpenAI Call" — makes the flow readable and gives you a clean place to add error handling later.

🎯 Sandbox Mission

Extend it on Power Automate

Add error handling with a Scope

Configure the error Scope: click the three dots → Configure run after → check "has failed". This is Power Automate's equivalent of a try/catch.

🧠 LLM Compatibility

Recommended

GPT-4o / GPT-4o-mini

OpenAI

FastLow costJSON mode

Best balance of speed/quality for email drafting and summarisation.

Claude 3.5 Sonnet

Anthropic

Long contextSafe

Excellent for long documents, nuanced tone, and multi-turn support.

Gemini 1.5 Pro

Google

1M contextMultimodal

Use when processing large documents or needing image + text in one step.

Mistral 7B / Mixtral

Mistral AI

Open weightSelf-host

Run locally via Ollama for zero API cost. Best for simple classification and drafts.

Llama 3.1 70B

Meta via Ollama

FreeLocal

Strong open model for production quality replies. Requires GPU server.

All models work with the n8n OpenAI, Anthropic, Google Gemini, and Ollama nodes. Swap models without changing workflow logic — only credentials and model name change.

🏆 Challenge: Customer Support Auto-Reply

Build a workflow that automatically replies to support emails using AI

A customer sends an email → your workflow reads it, uses GPT-4 to generate a helpful reply, and sends the response back. Build the correct node chain below.

📧 Read Email ✍️ Generate Text 📬 Send Email Connect them left → right

Triggers

🔗

Webhook

Trigger

📧

Read Email

Trigger

✍️

Generate Text

OpenAI / Claude

🏷️

Classify

AI Label

Output

📬

Send Email

Gmail / SMTP

📨

Send Message

Slack / Telegram

📤

Return Output

Response

Data

🔍

Extract Data

Parse fields

⚙️

Transform

Set / Code

↙️

Drag nodes from the left panel

Click another node to connect →

Automated YouTube Video Scheduling & AI Metadata Generation 🎬 — Automation University

📖 Overview

🏗️ Architecture

▶️ Animated Playback

Build Challenge

🔴 n8n Workflow

🏗️ Guided Build — n8n

🎯 Sandbox Mission

🟠 Make Scenario

🏗️ Guided Build — Make

🎯 Sandbox Mission

⚡ Zapier Zap

🏗️ Guided Build — Zapier

🎯 Sandbox Mission

🧩 Activepieces Flow

🏗️ Guided Build — Activepieces

🎯 Sandbox Mission

🔵 Power Automate Flow

🏗️ Guided Build — Power Automate

🎯 Sandbox Mission

📊 Live Stats

⚠️ Common Mistakes

🗺️ Active Quests

🏆 Achievements

Generate Text

What is Generate Text?

Core Concept

When to Use

Key Inputs

Key Output

Common Patterns

Webhook → Generate Text → Send Email

Read Email → Generate Text → Write Record

Read File → Generate Text → Return Output

Receive Webhook → Memory Retrieval → Generate Text → Reply

Loop Items → Generate Text → Aggregate Data

Compatible Nodes

Related Skills

Generate Text Workflows

Access Control for AI Agents (RBAC) using Airtable and Telegram

Access Real-time Binance Market Data with GPT-4o Formatting in Telegram

Action Plan: Utilizing OpenAI Model with Email Sources and Data Table

Ad Monitoring Tool with AI Model Names (GPT-4, Gemini)

Adaptive Email Auto-Responder with GPT-4, RAG and Human Feedback Loop

Adaptive RAG Strategy with Query Classification & Retrieval (Gemini & Qdrant)

Agentic Telegram AI Bot with LangChain Nodes and New Tools

AI Agent: Find the Right LinkedIn Profiles in Seconds

AI Agent: Scrape, Summarize & Save Articles to Notion (Gemini, Browserless)

AI Agent Chat — Conversational LLM with Memory

Automate Support Ticket Replies with Claude + Sentiment Detection

Daily Newsletter Generator from RSS Feeds using GPT-4o

Personalised Cold Email Writer from LinkedIn Profile Scrape

PDF Contract Summariser with GPT-4 + Slack Notification

Product Description Writer: Shopify SKUs → GPT-4 → CMS Update

Meeting Transcript → Action Items → Notion + Calendar Events

Practice Quiz

Build Challenge

🔴 n8n Workflow

🏗️ Guided Build — n8n

🎯 Sandbox Mission

🟠 Make Scenario

🏗️ Guided Build — Make

🎯 Sandbox Mission

⚡ Zapier Zap

🏗️ Guided Build — Zapier

🎯 Sandbox Mission

🧩 Activepieces Flow

🏗️ Guided Build — Activepieces

🎯 Sandbox Mission

🔵 Power Automate Flow

🏗️ Guided Build — Power Automate

🎯 Sandbox Mission

Related Workflows