Weekend Agent Challenge: Morning Brief Agent | Build on AWS

Learn how to build an automated Morning Brief AWS AI Agent using CloudFormation, Lambda, and Bedrock. Step-by-step guide and GitHub repository included.

Soumyadeep Mandal

Founder & Director @ OmniAi Global Solutions | AWS UG Kolkata Leader, AWS Community Builder

Published Jul 17, 2026

Last modified Jul 17, 2026

TL;DR / Key Takeaways

Fully Automated: This AWS AI Agent runs completely unattended every morning, generating a personalized briefing without manual triggers.
Free Tier Friendly: Built strictly using Amazon EventBridge, AWS Lambda, Amazon SNS, and Amazon Bedrock to keep costs practically non-existent.
Infrastructure as Code: The setup uses AWS CloudFormation for easy local deployment directly to your AWS account.

Introduction

Starting the day on the right foot often sets the tone for everything else. But honestly, waking up and immediately scrolling through scattered notes, news sites, or to-do apps is exhausting. The problem is that we spend way too much manual effort trying to organize our morning thoughts before we even have our coffee.

The simple solution is automation. Instead of you pulling information, what if a smart system pushed exactly what you needed directly to your phone or inbox?

That is exactly why I built this AWS AI Agent. In this post, I will walk you through my development process for creating a completely unattended morning briefing agent. I have updated the repository with the latest code, and I will show you exactly how the architecture works so you can deploy it yourself.

Vision & What the Agent Does
AWS Services Used & Architecture Overview
How You Built It: The Development Process
- Step 1: Setting up the Local Environment
- Step 2: The Python Logic (Lambda)
- Step 3: Infrastructure (CloudFormation)
What I Learned
Frequently Asked Questions
Conclusion

Vision & What the Agent Does

The core concept is to create a personal assistant that works while you sleep. The purpose of this agent is to generate a short, highly motivating morning brief and send it to your email right when you wake up. It solves the problem of morning decision fatigue.

Here is how it works from a user perspective:

The Trigger: I do not click anything. Every morning, the agent automatically wakes up based on a scheduled cron job.
The Unattended Action: It securely calls an AI model to generate a fresh, unique motivational quote and a quick productivity tip for the day.
The Report: It instantly formats this text and drops an email directly into my inbox, making it the first useful thing I read.

AWS Services Used & Architecture Overview

To keep this simple, reliable, and within the AWS Free Tier limits, I decided to go entirely serverless.

"Serverless architectures allow you to focus purely on the logic of your application without managing the underlying operating systems."

Here are the specific AWS services I used:

Amazon EventBridge Scheduler: This acts as our trigger. It runs the agent automatically on a set schedule without human interaction.
AWS Lambda: The core computing engine that runs our Python code.
Amazon Bedrock: The AI engine. We use a lightweight text model to ensure fast and cheap text generation.
Amazon SNS (Simple Notification Service): This handles the reporting aspect by taking the generated text and pushing it as an email alert.
AWS CloudFormation: This packages the entire architecture into a single template for easy deployment.

How You Built It: The Development Process

My development process was straightforward: draft the core logic in Python, test the Bedrock API connectivity, and then wrap the whole thing in a CloudFormation template. The main challenge was figuring out the exact IAM permissions required for Bedrock and SNS to talk to each other through Lambda. I overcame this by heavily restricting the policies in the YAML file to only allow the exact actions needed.

Here is how the project is structured.

Step 1: Setting up the Local Environment

You will need two primary files in your local project folder to make this work. I have organized them neatly in my GitHub repository. You need the application logic file and the infrastructure file.

Step 2: The Python Logic (Lambda)

This script is triggered by EventBridge, calls Amazon Bedrock for the AI content, and publishes the result to SNS.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
import json

import boto3

import os



.

.

.

.

[VISIT GITHUB FOR FULL CODE]

.

.

.

.

        

    except Exception as e:

        print(f"Error generating brief: {e}")

        raise e

Step 3: Infrastructure (CloudFormation)

This YAML file tells AWS exactly what to build. It sets up the roles, the Lambda function, the SNS topic, and the EventBridge rule in one go. Remember to change the placeholder email address to your actual email before deploying.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
AWSTemplateFormatVersion: '2010-09-09'

Description: Weekend Agent Challenge - Morning Briefing Agent



Resources:

  AgentSNSTopic:

    Type: AWS::SNS::Topic

    Properties:

      TopicName: MorningBriefingTopic



  AgentSNSSubscription:

    Type: AWS::SNS::Subscription

    Properties:

      TopicArn: !Ref AgentSNSTopic

      Endpoint: "YOUR_EMAIL@EXAMPLE.COM" # Change this to your email

      Protocol: email

.

.

.

.

[VISIT GITHUB FOR FULL CODE]

.

.

.

.

.

Outputs:

  TestUrl:

    Description: "URL to trigger the Morning Brief Agent Lambda function immediately for testing"

    Value: !GetAtt AgentLambdaFunctionUrl.FunctionUrl

Screenshots

AWS Cloudformation Deploy command using AWS CLI V2

Email Subscription confirmation by End-User

What I Learned

Participating in this challenge was a great experience. While I was already familiar with basic AWS services, integrating Amazon Bedrock directly via CloudFormation and standard Python libraries taught me a lot about AI infrastructure. I learned that you do not need heavy frameworks to build a reliable AI agent. A simple EventBridge trigger paired with properly scoped IAM roles is all it takes to build a secure, completely unattended workflow from scratch.

Frequently Asked Questions

What exactly is an AWS AI Agent?

An AWS AI Agent is an automated script or application running on Amazon Web Services that leverages artificial intelligence models to perform specific tasks, make decisions, or generate content without requiring manual human intervention.

How much does it cost to run this setup?

If you are operating within the AWS Free Tier, the costs for Lambda, EventBridge, and SNS for a single daily run are effectively zero. Amazon Bedrock charges per token, but a short daily prompt will cost fractions of a cent per month.

Can I change the time the agent runs?

Yes. You can easily modify the cron expression in the CloudFormation template (under the EventBridge settings) to trigger the agent at any specific time that suits your schedule.

Conclusion

Building your own AI assistant does not require massive budgets or complicated code. By using standard AWS Free Tier services like Lambda, EventBridge, and Bedrock, you can create a highly functional AWS AI Agent that runs entirely unattended. I hope this guide helps you realize how accessible building serverless AI applications can be. Take this foundation, update the code with your own ideas, and see what you can create.

Call to Action

Ready to deploy this yourself? I have uploaded all the updated codes to my GitHub repository. Clone the repo, review the files, and deploy your very own morning briefing agent today.

👉 View the Source Code on GitHub

Footer

About the Author:
Soumyadeep is a cloud developer passionate about simplifying complex AWS architectures. When not writing code, he enjoys exploring the latest advancements in generative AI and serverless computing.

Connect & Follow:
👉 X (Twitter) | LinkedIn | GitHub

Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.

Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article

Builder Center

Weekend Agent Challenge: Morning Brief Agent | Build on AWS

Soumyadeep Mandal

TL;DR / Key Takeaways

Introduction

Table of Contents

Vision & What the Agent Does

AWS Services Used & Architecture Overview

How You Built It: The Development Process

Step 1: Setting up the Local Environment

Step 2: The Python Logic (Lambda)

Step 3: Infrastructure (CloudFormation)

Screenshots

What I Learned

Frequently Asked Questions

Conclusion

Call to Action

Footer