AWS is managed with APIs, not clicks. This module shows how Python automates the cloud with Boto3 (read-only here), plus an intro to Infrastructure as Code with AWS CDK.

The scripts only read from AWS (list S3 buckets, list EC2 instances). Nothing is created, modified, or deleted unless you choose to run the optional create helpers.

AWS credentials configured locally:

aws configure   # or set AWS_ACCESS_KEY_ID / AWS_SECRET_ACCESS_KEY / AWS_DEFAULT_REGION

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

• s3_utilities.py — AWSUtils class: list buckets, optional create/upload
• aws_resource_report.py — list S3 + EC2 and write aws_report.json
• cdk_demo/ — an Infrastructure-as-Code example: a CDK stack with one S3 bucket

python aws_resource_report.py
python s3_utilities.py

These call real AWS APIs and need valid credentials — without them you'll get a NoCredentialsError / ClientError, which is expected.

Boto3 is great for reading state, reporting, and one-off automation. For creating/updating/deleting infrastructure, prefer IaC tools (Terraform, CloudFormation, CDK) — they track state and are repeatable. See cdk_demo/.

Write an AWS resource report:

1. Create a boto3.Session() and an S3 client, and list all bucket names.
2. List EC2 instances (instance id + state).
3. Print the report and save it to aws_report.json.
4. Bonus: wrap the calls in try / except (BotoCoreError, ClientError) so missing credentials print a friendly message instead of a traceback.

Read-only only — don't create or delete resources.