Block 1 - AI Agent
- Type / Role
- @n8n/n8n-nodes-langchain.agent - agent
- Config choices
- Version 1.8
This workflow is provided as-is. Please review and test before using in production.
Who is this for? This workflow is intended for WhatsApp admins, customer service, individual users or teams in businesses who want to automate their small business or medium to large businesses on ...
@n8n/n8n-nodes-langchain.agent, n8n-nodes-base.whatsapptrigger, n8n-nodes-base.if, n8n-nodes-base.whatsapp, @n8n/n8n-nodes-langchain.lmchatopenai, @n8n/n8n-nodes-langchain.memorybufferwindow, n8n-nodes-base.noop, n8n-nodes-base.set
This workflow is cataloged by N8N Workflows and links back to its original n8n.io source page by Ghufran Ridhawi.
Original n8n.io sourceThis workflow is intended for WhatsApp admins, customer service, individual users or teams in businesses who want to automate their small business or medium to large businesses on WhatsApp practically by automatically replying about your business products or services using Open AI capabilities on AI Agent. Based on an automated system there is a task to answer many customers or clients on WhatsApp, in reality answering one by one is very time consuming and tiring, especially if we have to compose sentences first or click on the template and send them. This is also a form of devotion to the community at n8n and the n8n company, as well as devotion to small businesses so that the reality is no longer tiring and able to answer the problems of reality that exist.
Easy explanation:
Whatsapp trigger here the "Input submissions" node is responsible for inputting chats from customers in the form of data. Then, it is sent to the next node, namely the if node, here there are two if nodes, namely "Signpost" and "Is text message?" responsible for forwarding customer chat input data and selecting if correct then it will be forwarded to the AI Agent and if wrong then there is a "Fallback Text" node. Once correct, the AI Agent will analyze with the OpenAI model capability and store it in Simple Memory. The AI Agent here has been given directions and commands on what it should be like, in the parameters section, the prompt contains a special task, the knowledge base of your business, and several points and rules that must be there. After that, it will be sent via the WhatsApp reply node which is addressed to the customer. After that, the Gmail Set will process it again to be addressed to the Send Gmail Notifications Node, which contains email address parameters and also the contents of the email as a notification of success or failure. Then, there are no operations after that after the Gmail node. And enjoy this workflow that works for you.
Complete what is in the node as stated in the notes column. You need a "Credential Account" on each WhatsApp node, then the OpenAI node and the Gmail Node. If you already have an account, just connect it, and if not, create it first, you can register it by following the guide from n8n, it makes it very easy. Because here it has been set up neatly and cleanly like in the if node. Next, you need to add your business to the AI Agent node, there are parameters and a prompt section, I have made a sample here to make it easier for you. You can immediately adjust it to your business or service or product, later what communication or sentence you want to convey to customers on WhatsApp. After that, don't forget to add an email address to the Gmail node for notification of success or failure of each customer from the running of this system. Save and run, test the workflow and activate the workflow. This workflow is ready to use.
As a reminder:
Must be set in nodes such as your business description, also according to the conditions of your business, services, products, so that the AI Agent matches your business knowledge base Must have (if not, make sure you have registered) in each "Credential Account" by following the guide on how to create it n8n the guide is very complete Don't forget to save, and make sure the workflow is active.
## How to customize this workflow to your needs You can immediately set up your business knowledge base in the nodes, so that the accuracy is also high when carrying out tasks and answering them.
This catalog entry is organized from the workflow JSON. The node-level section below shows the executable blocks available for review before importing the template.
| Workflow | WhatsApp AI customer service bot with GPT-4o-mini and Gmail alerts |
|---|---|
| Complexity | advanced |
| Nodes | 22 |
| Categories | Support Chatbot, AI Chatbot |
| Author | Ghufran Ridhawi |
| Published | 28 May 2025 |
Use the JSON export at /data/workflows/4456/4456.json as the source template for this automation.
Open n8n, import the downloaded JSON, and review each node before activating the workflow.
Replace placeholder credentials, API keys, webhook URLs, account IDs, and environment-specific values with your own settings.
Run the workflow manually or in a staging workspace, inspect node output, and confirm downstream systems receive the expected data.
Enable the workflow only after testing, then monitor executions, errors, and rate limits during the first production runs.
Review imported nodes carefully before activation. This catalog entry is intended to help you inspect the workflow structure, understand required services, and find related templates faster.
Node names, credentials, schedules, webhook paths, and external service limits may need adjustment for your workspace.
Who is this for? This workflow is intended for WhatsApp admins, customer service, individual users or teams in businesses who want to automate their small business or medium to large businesses on ...
Review the workflow JSON, configure any required credentials in n8n, and test the automation in a safe workspace before using it in production.
Yes. Use the block-by-block analysis and the downloadable JSON to inspect each node, then adjust credentials, prompts, schedules, filters, or destinations for your Support Chatbot, AI Chatbot use case.