Demo Create an S3 bucket with Python and the CDK

This hands‑on guide demonstrates how to use the AWS CDK (Cloud Development Kit) with Python to create an S3 bucket. Follow the sequence below: install prerequisites, initialize a Python CDK project, add an S3 bucket construct, synthesize the CloudFormation template, bootstrap (if required), deploy the stack, verify the resources, and clean up.

Prerequisites

Node.js (includes npm) — required for the CDK CLI.
AWS CLI installed and configured with credentials and a default region.

Verify AWS connectivity (optional):

aws sts get-caller-identity

Install the CDK CLI

Install the CDK CLI globally:

npm install -g aws-cdk

Verify installation:

cdk --version
cdk --help

Important CDK CLI commands

Command	Purpose
`cdk ls` / `cdk list`	Lists stacks in the app
`cdk synth`	Synthesizes and prints the CloudFormation template
`cdk bootstrap`	Deploys CDK bootstrap resources into an AWS environment
`cdk deploy`	Deploys the stack(s) to your AWS account
`cdk diff`	Compares the local stack with the deployed stack
`cdk destroy`	Removes the deployed stack(s)

Initialize a new Python CDK project

Create a starter project:

cdk init sample-app --language python

This generates the standard Python CDK layout:

Virtual environment placeholder (we’ll create it next)
source.bat (Windows helper)
requirements.txt (Python dependencies)
README (usage and test instructions)
cdk.json (CDK config)
app.py (CDK app entrypoint)
A module folder with a generated Stack class

Create and activate the Python virtual environment

macOS / Linux:

python -m venv .venv
source .venv/bin/activate

Windows (PowerShell):

python -m venv .venv
.venv\Scripts\Activate.ps1

Note: The generated source.bat replicates source .venv/bin/activate behaviour on Windows by calling .venv\Scripts\activate.bat. Example source.bat (for reference):

@echo off

rem The sole purpose of this script is to make the command
rem     source .venv/bin/activate
rem (which activates a Python virtualenv on Linux or Mac OS X) work on Windows.
echo Executing .venv\Scripts\activate.bat for you
.venv\Scripts\activate.bat

Install Python dependencies

After activating the virtual environment, install dependencies:

pip install -r requirements.txt

A typical requirements.txt for CDK v2:

aws-cdk-lib==2.101.0
constructs>=10.0.0,<11.0.0

Project entrypoint: app.py

The generated app.py bootstraps the CDK app and instantiates the stack:

#!/usr/bin/env python3

import aws_cdk as cdk

from cdk.cdk_stack import CdkStack

app = cdk.App()
CdkStack(app, "CdkStack")

app.synth()

Generated sample stack (example)

The sample stack demonstrates how CDK resources are defined (the generated template often includes SQS/SNS examples). For this tutorial you can comment those out and add an S3 bucket instead:

from constructs import Construct
from aws_cdk import (
    Duration,
    Stack,
    aws_iam as iam,
    aws_sqs as sqs,
    aws_sns as sns,
    aws_sns_subscriptions as subs,
)

class CdkStack(Stack):

    def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None:
        super().__init__(scope, construct_id, **kwargs)

        queue = sqs.Queue(
            self, "CdkQueue",
            visibility_timeout=Duration.seconds(300),
        )

        topic = sns.Topic(
            self, "CdkTopic"
        )

        topic.add_subscription(subs.SqsSubscription(queue))

For this lesson we will not use the SQS/SNS resources — comment them out and add the S3 bucket as shown below.

Create an IAM user for deployment

CDK uses the same AWS credentials mechanism as the AWS CLI. Create an IAM user with programmatic access and generate an access key/secret key. Prefer least privilege rather than Administrator access.

Best practice: avoid using AdministratorAccess for deployment agents. Create a policy that grants only the permissions required by your CDK app.

When creating the IAM user you will see screens like this:

A screenshot of the AWS Management Console's IAM "Create user" page showing a form to enter a username and optional console access. The left sidebar shows setup steps and there's a highlighted "Next" button at the bottom right.

Attach the required policy (or AdministratorAccess for quick demos):

A screenshot of the AWS IAM console showing the "Permissions policies" list while creating a user, with options to add or attach policies directly and a table of AWS-managed policy names (e.g., AdministratorAccess). The page includes search/filter controls and pagination for browsing available policies.

When creating an access key, you’ll see best-practice guidance:

Screenshot of the AWS IAM "Create access key" page showing "Access key best practices & alternatives" with selectable use-case options (CLI, local code, AWS compute, third-party, etc.). The bottom-right has "Cancel" and an orange "Next" button.

Configure the keys locally:

aws configure
# AWS Access Key ID [None]: <YOUR_ACCESS_KEY_ID>
# AWS Secret Access Key [None]: <YOUR_SECRET_ACCESS_KEY>
# Default region name [None]: us-east-1
# Default output format [None]: json

Add an S3 bucket to the stack

Open your stack file (for example cdk/cdk_stack.py) and:

Add aws_s3 to the import block:

from aws_cdk import (
    Duration,
    Stack,
    aws_iam as iam,
    aws_sqs as sqs,
    aws_sns as sns,
    aws_sns_subscriptions as subs,
    aws_s3 as s3,
)

Comment out sample SQS/SNS code (if present) and add a simple S3 bucket:

class CdkStack(Stack):

    def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None:
        super().__init__(scope, construct_id, **kwargs)

        # Example: keep SQS/SNS commented out if not needed
        # queue = sqs.Queue(
        #     self, "CdkQueue",
        #     visibility_timeout=Duration.seconds(300),
        # )

        # Create a simple S3 bucket:
        bucket = s3.Bucket(self, "MyKodeKloudCdkBucket")

Notes:

The logical ID "MyKodeKloudCdkBucket" is used by CDK when synthesizing resources. CDK generates a unique physical bucket name by default for global uniqueness.
To use a fixed bucket name, set the bucket_name property (ensure global uniqueness and S3 naming rules).
Configure encryption, versioning, lifecycle rules, or removal_policy via constructor arguments as needed.

Synthesize the CloudFormation template

Generate the CloudFormation template:

cdk synth

This prints the CloudFormation YAML/JSON that CDK will deploy. Example snippet from cdk synth (trimmed):

Resources:
  MykodekloudcdkbucketB75EFD9A:
    Type: AWS::S3::Bucket
    UpdateReplacePolicy: Retain
    DeletionPolicy: Retain
    Metadata:
      aws:cdk:path: CdkStack/My-kodekloud-cdk-bucket/Resource

Bootstrap the environment (if required)

If you haven’t bootstrapped this account/region for CDK v2, cdk deploy will ask you to run:

cdk bootstrap

Bootstrapping deploys supporting resources (an S3 staging bucket, roles, etc.). If bootstrap fails because of resource conflicts (e.g., an S3 bucket with the same name exists), resolve the conflict and re-run cdk bootstrap.

Preview and deploy

Preview changes:

cdk diff

If the diff matches your intent, deploy:

cdk deploy

During deploy you’ll see CloudFormation progress. Example trimmed output:

CdkStack | 0/3 | 11:54:30 PM | CREATE_IN_PROGRESS | AWS::S3::Bucket | My-kodekloud-cdk-bucket
CdkStack | 2/3 | 11:54:51 PM | CREATE_COMPLETE    | AWS::S3::Bucket | My-kodekloud-cdk-bucket
CdkStack | 3/3 | 11:54:51 PM | CREATE_COMPLETE    | AWS::CloudFormation::Stack | CdkStack

✅ CdkStack

✨ Deployment time: 33.21s

Verify in the console

After deployment, check the S3 console. The created bucket appears in the Buckets list (region shown):

A screenshot of the Amazon S3 web console showing a list of S3 buckets with their names, AWS regions, access settings, and creation dates, plus the left navigation panel and a "Create bucket" button. Five buckets are listed, mostly in the US East (N. Virginia) region.

The CDK deploys your construct via CloudFormation. You can inspect the stack in the CloudFormation console to see resources and status:

Screenshot of the AWS CloudFormation console showing stacks on the left and the Resources tab for a stack named "CdkStack" on the right. It lists two resources — My-kodekloud-cdk-bucket and CDKMetadata — both with status CREATE_COMPLETE.

Post-deploy checks

Run cdk diff again to confirm no outstanding changes:

cdk diff
# Output: There were no differences

Destroying the stack

When finished, remove the stack:

cdk destroy

You will be prompted to confirm deletion.

If the S3 bucket contains objects, CloudFormation cannot delete the bucket and cdk destroy will fail. Either empty the bucket first (console, CLI, or SDK), or configure the bucket for development/testing with removal_policy=cdk.RemovalPolicy.DESTROY and auto_delete_objects=True to enable automated deletion. Do not enable auto_delete_objects in production without understanding the consequences.

After destroy completes (or after emptying and re-running destroy), verify the CloudFormation stack and S3 bucket are removed in the console.

Conclusion

You learned how to:

Initialize a Python CDK project
Create and activate a virtual environment
Add an S3 bucket construct to the stack
Synthesize the CloudFormation template
Bootstrap the environment (if required)
Deploy and verify the bucket
Destroy resources when finished

The CDK lets you define infrastructure as code, synthesize a CloudFormation template, and deploy predictable infrastructure. Explore additional CDK constructs and best practices as you build more sophisticated stacks.

Links and references

AWS CDK Developer Guide: https://docs.aws.amazon.com/cdk/latest/guide/home.html
AWS CDK API Reference: https://docs.aws.amazon.com/cdk/api/v2/
AWS CloudFormation: https://learn.kodekloud.com/user/courses/aws-cloud-formation
AWS IAM basics: https://learn.kodekloud.com/user/courses/aws-iam
Amazon S3: https://learn.kodekloud.com/user/courses/amazon-simple-storage-service-amazon-s3

Introduction

Services - Networking

Services - Storage

Services - Compute

Services - Database

Services - Application Integration

Services - Data and ML

Services - Migration and Transfer

Services - Management and Governance

Services - Security

Bringing it all together

Designing for Security

Designing for Reliability

Designing for Performance

Designing for Cost-Optimization

Applying your Design Skills

Demo Create an S3 bucket with Python and the CDK

Prerequisites

Install the CDK CLI

Important CDK CLI commands

Initialize a new Python CDK project

Create and activate the Python virtual environment

Install Python dependencies

Project entrypoint: app.py

Generated sample stack (example)

Create an IAM user for deployment

Add an S3 bucket to the stack

Synthesize the CloudFormation template

Bootstrap the environment (if required)

Preview and deploy

Verify in the console

Post-deploy checks

Destroying the stack

Conclusion

Links and references

Watch Video

​Prerequisites

​Install the CDK CLI

​Important CDK CLI commands

​Initialize a new Python CDK project

​Create and activate the Python virtual environment

​Install Python dependencies

​Project entrypoint: app.py

​Generated sample stack (example)

​Create an IAM user for deployment

​Add an S3 bucket to the stack

​Synthesize the CloudFormation template

​Bootstrap the environment (if required)

​Preview and deploy

​Verify in the console

​Post-deploy checks

​Destroying the stack

​Conclusion

​Links and references

Watch Video

Prerequisites

Install the CDK CLI

Important CDK CLI commands

Initialize a new Python CDK project

Create and activate the Python virtual environment

Install Python dependencies

Project entrypoint: app.py

Generated sample stack (example)

Create an IAM user for deployment

Add an S3 bucket to the stack

Synthesize the CloudFormation template

Bootstrap the environment (if required)

Preview and deploy

Verify in the console

Post-deploy checks

Destroying the stack

Conclusion

Links and references