A CLI tool for working with IaC in AWS Proton.

AWS Proton provides a self-service deployment service with versioning and traceability for your IaC templates. The Protonizer CLI tool lets you scaffold out new Proton templates from scratch as well as allows you take your existing IaC (infrastructure as code) templates and modules and bring them into AWS Proton to scale them out across your organization.

Note that this is an experimental project and currently supports generating Proton templates based on existing Terraform and CodeBuild provisioning. The tool also currently only supports primitive HCL data types such as strings, numbers, bools, and lists of primitive types. This is currently aligned with the Proton schema types that are supported by the Proton console.

To install the protonizer CLI tool, you can download the latest binary release for your platform and architecture.

Protonizer can scaffold out all of the files you need to build a Proton template. It can them register and publish the template using the publish command.

Protonizer can also parse your existing Terraform modules and generates Proton templates with schemas based on your input and output variables. It also outputs manifest.yml files that will run terraform apply within a Proton-managed environment.

The new command scaffolds out new Proton templates from scratch.

Terraform

protonizer new \
  --name my_template \
  --type service \
  --provisioning codebuild --tool terraform \
  --terraform-remote-state-bucket my-s3-bucket \
  --publish-bucket my-s3-bucket \
  --out ~/proton/templates

tree
.
| |____my_template
| | |____v1
| | | |____proton.yaml
| | | |____schema
| | | | |____schema.yaml
| | | |____instance_infrastructure
| | | | |____outputs.tf
| | | | |____main.tf
| | | | |____output.sh
| | | | |____manifest.yaml
| | | | |____install-terraform.sh
| | | | |____variables.tf

CloudFormation

protonizer new \
  --name my-template \
  --type environment \
  --provisioning awsmanaged \
  --out ~/proton/templates \
  --publish-bucket my-s3-bucket

tree
.
| |____my-template
| | |____v1
| | | |____proton.yaml
| | | |____schema
| | | | |____schema.yaml
| | | |____infrastructure
| | | | |____cloudformation.yaml
| | | | |____manifest.yaml

The protonize command can generate and publish a CodeBuild provisioning template based on an existing Terraform module.

protonizer protonize \
  --name my_template \
  --type environment \
  --provisioning codebuild --tool terraform \
  --terraform-remote-state-bucket my-s3-bucket \
  --dir ~/my-existing-tf-module \
  --out ~/proton/templates \

template source outputted to ~/proton/templates/my_template
done

protonizer protonize \
  --name my_template \
  --type service \
  --compatible-env env1:1 --compatible-env env2:1 \
  --provisioning codebuild --tool terraform \
  --terraform-remote-state-bucket my-s3-bucket \
  --dir ~/my-existing-tf-module \
  --out ~/proton/templates \
  --publish-bucket my-s3-bucket \
  --publish

template source outputted to ~/proton/templates/my_template
published my_template:1.0
https://us-east-1.console.aws.amazon.com/proton/home?region=us-east-1#/templates/services/detail/my_template
done

The publish command registers and publishes a template with AWS Proton. Just add a proton.yaml file to your project and run protonizer publish. This is alternative to Proton's Template sync feature, useful for local development or for Git providers that aren't supported.

proton.yaml

name: my_template
type: environment
displayName: My Template
description: "This is my template"
publishBucket: my-s3-bucket

publish using yaml file

protonizer publish
published my_template:1.0
https://us-east-1.console.aws.amazon.com/proton/home?region=us-east-1#/templates/environments/detail/my_template

proton.yaml

name: my_template
type: service
displayName: My Template
description: "This is my template"
compatibleEnvironments:
  - env1:3
  - env2:4
publishBucket: my-s3-bucket

publish using yaml file

protonizer publish
published my_template:1.0
https://us-east-1.console.aws.amazon.com/proton/home?region=us-east-1#/templates/services/detail/my_template

or specify file name

protonizer publish -f file.yml
published my_template:1.0
https://us-east-1.console.aws.amazon.com/proton/home?region=us-east-1#/templates/environments/detail/my_template

Note that this can also be done inline with the protonize --publish command.

To avoid conflicts, if you have variables in your source templates with reserved names in Proton (i.e., name and environment), they will be removed as template input variables and instead be sourced from proton metadata.

If the source terraform module has an input variable named name, it will be supplied by the name of the proton environment rather than by template specific input.

If the source terraform module has a variable named name, it will be set to the name of the service and the service instance with a - (dash) in between. If the source terraform module has a variable named environment, it will be set to the service instance's environment name.

For example, when creating a service named sales-api and a service instance named dev associated with a proton environment named dev, the Terraform module will get passed the following values:

name = "sales-api-dev"
environment = "dev"

• Go 1.20
• Install pre-commit
• Run pre-commit install to setup git hooks

 Choose a make command to run

  vet           vet code
  test          run unit tests
  build         build a binary
  autobuild     auto build when source files change
  dockerbuild   build project into a docker container image
  start         build and run local project
  deploy        build code into a container and deploy it to the cloud dev environment
  xplat         multiplatform build