AWS CDK Script Generator
Pick the AWS resources you need and get a ready-to-deploy CDK v2 stack in TypeScript or Python, with permissions and outputs wired up.
Resources
- Run
cdk init appfirst, then replace the generated stack and app files with these. - Buckets and tables default to
RETAINso a stack delete never destroys data; tick “Delete on stack destroy” only for throwaway environments. - When Lambda is combined with S3 or DynamoDB, the function gets the resource name as an environment variable and read/write permission.
- Review
cdk diffbefore every deploy; nothing here touches AWS until you run it yourself.
Your Data Never Leaves Your Device
Every tool runs entirely in your browser. Nothing you type is uploaded, stored, or logged on our servers.
The AWS CDK Script Generator turns a few form choices into a complete CDK v2 stack, an app entry file and the CLI commands to bootstrap and deploy it, in TypeScript or Python.
Pick from S3, DynamoDB, Lambda, API Gateway, SQS, SNS and VPC. The resources reference each other correctly by construction: a Lambda function receives the table and bucket names as environment variables and is granted read/write access, the API is a proxy integration on that function, and stack outputs expose the names and URL you will need after deploy. Stateful resources default to RETAIN so deleting a stack does not delete data.
FAQ
AWS CDK v2. Imports come from the single aws-cdk-lib package (TypeScript) or aws_cdk (Python), not the per-service @aws-cdk/* packages used by v1.
No. It only produces source code in your browser. Nothing is sent anywhere and no credentials are needed. You deploy by running cdk deploy yourself.
With RETAIN, deleting the stack leaves your data in place. Tick 'Delete on stack destroy' only for disposable environments; for buckets this also adds autoDeleteObjects so the bucket can be emptied and removed.
Run cdk init app with the same language, then replace the generated stack file and app entry file with the generated ones. Put your Lambda handler in the code directory you chose, then run cdk bootstrap (once), cdk diff and cdk deploy.
The generated API is a LambdaRestApi that proxies every request to your function, so it needs a Lambda function to point at.
Yes. The output is plain CDK code, so treat it as a starting point and edit it freely, for example to add IAM policies, event sources or additional stacks.