N8N + WhatsApp GRÁTIS: Crie seu Agente de IA sem gastar NADA! (Tutorial fácil para iniciantes 2025)
How to Create a Free WhatsApp Integration Agent
Introduction to the Project
- The video aims to teach viewers how to create an integrated WhatsApp agent using N8N, WAHA API, and Redis database for free.
- It emphasizes that no payment is required for any of the solutions presented, making it accessible for beginners in automation.
Setting Up Docker
- Viewers are instructed to download and install Docker Desktop from its official website, with options available for MacOS, Windows, and Linux.
- After installing Docker, users need to download a specific file called
dockercomposeand place it in a newly created folder named "N8N waha local."
- A terminal command (
Docker-compose up -d) is provided to automatically set up N8N, WAHA, and Redis within Docker. This command initiates the necessary services.
Configuring WAHA
- Users are guided on how to access the WAHA dashboard through a link in Docker and connect their WhatsApp by scanning a QR code with their mobile device.
- Once connected successfully, users can proceed with configuring N8N for further automation tasks.
Setting Up N8N
- Instructions include entering personal details (email, name) when prompted by N8N after accessing it via Docker. Users should select options indicating they are not using it for work purposes.
- A free activation key is obtained via email which unlocks additional features in N8N; this key must be entered into the settings under "integration key."
Creating Workflows in N8N
- Users learn how to initiate a new workflow titled "WhatsApp" and add a webhook trigger configured as POST with a specified URL matching previous configurations.
- The process involves setting up event handling by deselecting unnecessary events while ensuring message events remain active for processing incoming messages from WhatsApp.
Data Handling in Workflows
- A new node labeled "set" is added to filter relevant data from incoming webhooks; specific fields such as session ID and message body are selected for processing.
Messenger Event Configuration
Setting Up the Messenger Event
- The discussion begins with different routes for various event types, focusing on the Messenger event for now. If the event equals "message," it confirms that the setup is correct.
- The configuration of an AI agent is introduced, where a prompt is defined to guide the agent's responses based on incoming messages.
Creating and Configuring the AI Agent
- A simple prompt example is provided: "You are a tour guide who knows everything about Foz do Iguaçu and helps travelers create personalized itineraries." This sets up the context for the AI agent.
- Instructions are given on obtaining a free API key from Google Gemini by logging into Google AI Studio and creating a new key.
Connecting to Google Gemini
- After copying the API key, it’s pasted into N8N to establish a successful connection. The model used will be Gemini 2.0 Flash, which offers improved performance over version 1.0 Pro.
Adjusting Model Parameters
- An optional parameter called sampling temperature can be adjusted to control how creative or "crazy" the AI's responses may be; default is set at 0.4.
Memory Configuration in N8N
Adding Chat Memory
- A Redis chat memory node is added using pre-configured settings from Docker Compose, requiring specific credentials like password and host adjustments.
Session ID Setup
- The session ID configuration involves setting it to define below and linking it with chat ID data while specifying session duration (3600 seconds).
Finalizing Workflow for WhatsApp Integration
Sending Messages via WhatsApp
- A new node is added to send messages through WhatsApp after clearing existing data fields and selecting appropriate credentials.
Testing Message Delivery
- The workflow tests message delivery by linking session IDs and message IDs correctly; successful visual confirmation appears in WhatsApp.
Demonstrating Memory Functionality
Interaction Example with Memory Feature
- An interaction showcases memory functionality where user input ("My name is Guilherme") leads to accurate recall when queried about their name.
Limitations of Local N8N Installation
Constraints on Automation Capabilities
WhatsApp Automation and New Course Announcement
Overview of WhatsApp Automation Installation
- The installation discussed is specifically for automating WhatsApp using the Ah tool, with no additional features included.
Introduction to New Course
- A new course has been launched that teaches how to create AI agents using N8N, Ah, or Evolution both locally and on a VPS.
- The course covers essential fundamentals of N8N along with various bonuses that will be added over time.
Additional Learning Opportunities
- Future content will include topics such as website and app creation, API integrations, API development, database management, and more.
Promotional Offer
- The course is currently available at a promotional launch price during May.
VPS Subscription Recommendation