“We are at the most interesting time to be alive of anytime in history!”

Elon Musk

Elon Musk said this during the Grok 4 release, and I could not agree more. Just 5 years ago, building chatbots was too slow, chancy, and frankly synthetic.

Now we are not just talking about chatbots; they have evolved. We are now building LLM-powered chatbots that think, act, and grow, which are called Agents.

In this article, we will explore how to build Agents by using zero code and n8n, but first, let’s see what n8n is.

What is n8n?

Reference

n8n is an automation platform. But after AI, automation platforms will also give you a chance to integrate AI models into their workflow.

Let’s explore the n8n workflow that we are using in one of our solutions, the startup idea generator.

Here is the typical n8n screen.

Reference

Workflows

Do you see workflows? These are the places where you create automations. Please check the screenshot below.

Now, let’s click Agent-Startup-idea-generator, which we are using as a chatbot, for our next solution. Here, you can test the agent; it is live.

As you can see above, it is pretty straightforward.

Trigger

In step 1, you have to define the trigger. In chatbots, it will be a message. But there are other triggers like Webhook, schedule, or even manual. Here are all of your options.

Triggers , n8n

If you want to stay ahead in AI with hands-on projects, sharp insights, and weekly tools, subscribe to our Substack.

Agent

Next step, AI Agent. Click on the AI from the screen below, and you will see the AI Agent.

AI Agent

In an AI agent, there are three different things you can use.

• Chat Model
• Memory
• Tool

Memory

As a memory, you can use Simple Memory, which is free on n8n, and you don’t need to add credentials; however, there are other options here to discover.

Memory- n8n

Chatbot

This is pretty simple, like ChatGPT, Claude, Gemini, or Grok. You have to add your API to the platform, so you can select one of these models.

AI Models- n8n

Tool

Next, we'll add the tool, which is where things get interesting. Here, you are going to add superpowers to your agent. Let’s start simple, let’s add a web browsing feature, like Serp API.

But how would your agent know how to use this? Of course, you have to define the prompt, but where? Click on the AI Agent and define the system message.

Here, you can tweak the agent as you like, allowing it to use the tool you've defined.

AI News Agent

Now let’s create an AI News Agent that will give you the AI news and browse about it when you want it.

AI News Agent — n8n

• Step 1- Select the trigger
• Step 2- AI Agent
• Step 3- Add Chatbot
• Step 4- Simple Memory
• Step 5- Adding Tools — SerpAPI & RSS-1

Now you know how to do Steps 1 to 4. Let’s see how to add RSS-1. Click on the “+” button under Tool and add the RSS read tool.

RSS Read Tool.

Go to Google Alerts and create a Google Alert about AI, take the RSS, and put this URL.

Good, now let’s tweak the prompt like this, after clicking on the AI Agent as a system message.

Here is the prompt.

You are an autonomous assistant triggered on a schedule.

Your task is to:

1. Collect the latest AI news from RSS-1.
2. Use browsing when needed.

Final Adjustments

Click on the first node: “When chat message received.”, and make sure to select “Make Chat Publicly Available”.

Good, now let’s visit the ChatURL. But before that, make sure your workspace is available.

Workflow Status

Testing AI News Agent

Now let’s test it.

As you can see, you will see daily AI news and even summarize it with 2 prompts.

Final Thoughts

In this article, we have created a simple n8n agent, which will give you daily AI news and even summarize it. You can add nodes to send this message to your Slack or save them to Excel.

Of course, tracking AI news should not be that hard, so if you want to follow AI News easily, visit our platform. See you there!

Here are the free resources.

Here is the ChatGPT cheat sheet.

Here is the Prompt Techniques cheat sheet.

Here is my NumPy cheat sheet.

Here is the source code of the “How to be a Billionaire” data project.

Here is the source code of the “Classification Task with 6 Different Algorithms using Python” data project.

Here is the source code of the “Decision Tree in Energy Efficiency Analysis” data project.

Here is the source code of the “DataDrivenInvestor 2022 Articles Analysis” data project.

“Machine learning is the last invention that humanity will ever need to make.” Nick Bostrom

Here are paid resources.

Here you can create your start-up ideas.

Here is our Substack page.