n8n Fundamentals

Learning Objectives

1. Lesson 1

   Introduction

   • Explain what n8n is and how it differs from doing a task by hand
   • Set up and navigate the n8n editor — canvas, nodes, and connections
   • Build and run a simple end-to-end workflow, from trigger to output
   • Read a workflow run to see what happened at each step
2. Lesson 2

   Your First Workflow

   • Identify the core building blocks of a workflow: triggers, nodes, and the data that flows between them
   • Build a form-to-spreadsheet workflow from scratch
   • Reshape and clean up incoming data as it moves through a workflow
   • Check a run's output to confirm the workflow did what you expected
3. Lesson 3

   Fix the Flaw

   • Explain why overwriting a file on every run causes lost data
   • Combine new data with what's already there instead of replacing it
   • Add a duplicate check so the same record isn't entered twice
   • Rebuild a flawed workflow into one that's safe to run repeatedly
4. Lesson 4

   Batch Processing, Live

   • Trigger a workflow on a schedule instead of running it by hand
   • Process a batch of records in one run instead of one at a time
   • Route incomplete or invalid records aside for review
   • Build and activate a complete pipeline, start to finish
5. Lesson 5

   App Nodes

   • Connect a workflow to real apps — Google Sheets and Gmail
   • Read from and write to a spreadsheet directly, without import/export steps
   • Send an automated email as part of a workflow
   • Set up an alert that notifies you when a workflow fails
6. Lesson 6

   Teaching a Workflow to Read

   • Explain what adding AI to a node changes about what a workflow can do
   • Prompt AI within a workflow to return information in a usable, structured format
   • Turn an AI response into individual fields the rest of the workflow can use
   • Build a workflow that reads unstructured input and pulls out what matters
7. Lesson 7

   Building an Agent

   • Explain the difference between a workflow that reads and one that decides for itself
   • Give an AI agent a tool it can call on its own, plus memory to hold a conversation
   • Explain the Agentic Loop — Listen, Plan, Execute, Evaluate — and the four parts of an agent: agent, model, tools, memory
   • Build a chat agent that looks up real data before it answers, and know its limits
8. Lesson 8

   Building Workflows with AI

   • Explain what MCP is and how it connects an AI assistant to n8n
   • Set up Claude to build and edit workflows directly
   • Describe a workflow in plain language and have Claude build it
   • Verify an AI-built workflow is correct before running it