Build an AI Agent to Summarise LinkedIn Posts with Relay.app (No Code Required)

Build an AI Agent to Summarise LinkedIn Posts with Relay.app (No Code Required)

Zach Measures

AI agents are THE big thing for 2025.

How useful are they though?

 It turns out.... quite useful!

In just a few minutes, you can build an AI agent using Relay.app that fetches LinkedIn posts, analyzes them with AI, and sends you clean, summarised insights directly to your inbox — all without writing a single line of code.

This is obviously just one use case! 

 Before we get too far into this, it's important to actually sketch out what AI agents actually are. 

According to Chat GPT: 

"An AI agent is a computer program or system that can asses elements of its environment, make decisions, and take actions to achieve specific goals."

Okay.

Sounds impressive. But what does this actually mean....Is this a Chatbot or what?

You can think of these technologies broken down into three key areas or modalities: 

  1. Chatbots - LLMs like ChatGPT or Claude. These are tools which respond to human inputs in text, images or other data sets. 
  2. Copilots - (Grammarly or Cursor) again, these work with human inputs but provide active prompts and suggestions based on inputs. 
  3. AI Agents. These are different in the sense they do autonomous work in the background!

So, given that AI agents can autonomously work on your behalf, without further human inputs, their value becomes clear....

 The below is my first foray into creating an agent and I wouldn't have been able to do it without the excellent and easy user interface of relay.app.

 So let's get this bad boy started: 

✅ What this will do:

By the end of this post, you'll know how to:

  1. Set up a scheduled trigger to run your agent automatically
  2. Fetch LinkedIn posts from chosen profiles
  3. Use Chat GPT / Claude 3.5 Sonnet to summarize and extract key info
  4. Configure automated email reports so insights come to you

⏱️ The 3-Minute Setup (Step-by-Step)

Step 1: Go to Relay.app

Step 2: Create an account which is free using an email. I used a "burner" email as you can link it to your Google Drive.

Step 3. Search for LinkedIn in the preset templates 

Start to follow their easy prompts!

Start to follow their easy prompts!

TIP: Creating a Scheduled Trigger In Relay, start with a trigger: “Run every day at 8 AM” — or any frequency you prefer. This tells your agent when to begin working.

TIP: Simply add in the list of LinkedIn URL's for your industry profiles

TIP: Simply add in the list of LinkedIn URL's for your industry profiles

<br>

<br>

Finally, send the email to yourself! 

That's it! 

You now have a fully automated LinkedIn research assistant — ready to deliver curated insights every day.

💡 Bonus Ideas

Once you’ve got the basics working, here are a few extra enhancements:

  1. Tag sentiment or categorize posts (e.g., job updates, product announcements, thought leadership)
  2. Add Slack integration if your team wants real-time updates
  3. Connect to Airtable or Notion to store historical summaries

Try it out and let your new AI agent do the heavy lifting.