A command-line tool for managing AWS CloudFormation stacks as code.
📚 Read the user documentation for tutorials, how-to guides, explanations, and detailed reference material.
Stackaroo simplifies CloudFormation stack management through declarative YAML configuration, allowing you to define your infrastructure once and deploy it consistently across multiple environments. It provides comprehensive parameter management with support for literal values, dynamic stack output resolution, and cross-region references. Features include dependency-aware deployment ordering, integrated change previews, template validation, and real-time event streaming during stack operations.
- Deploy the same templates across multiple contexts
- Different AWS regions and parameters per context
- Define stack dependencies with
depends_on - Automatic deployment ordering
- Shows template, parameter, tag, and resource changes in a unified diff format (similar to
git diff). - Uses the AWS ChangeSet API to identify which resources will be created, modified, deleted, or replaced.
- Highlights resources that require replacement and uses the same format as the dedicated
diffcommand.
- Displays status, creation time, last update, description, parameters, outputs, and tags for a deployed stack.
- Retrieves current data directly from AWS CloudFormation in a clean, consistently formatted layout.
- Validates templates against the CloudFormation API in the target region without deploying, catching syntax errors and invalid resource types.
- Supports single-stack and whole-context validation, making it suitable for pre-deployment checks in CI/CD pipelines.
Validate templates early in your development workflow:
# Validate a single stack's template
stackaroo validate dev vpc
# Validate all stacks in a context
stackaroo validate productionThe validation command provides immediate feedback on template errors without requiring actual deployment, making it ideal for development workflows and continuous integration pipelines. It processes templates through the same resolution pipeline as deployment, including Go template processing, ensuring validation matches what will actually be deployed.
Stackaroo provides a comprehensive parameter system supporting multiple resolution types:
Direct string values defined in configuration:
parameters:
Environment: production
InstanceType: t3.medium
Port: "8080"Pull values dynamically from existing CloudFormation stack outputs:
parameters:
VpcId:
type: stack-output
stack: networking
output: VpcId
DatabaseEndpoint:
type: stack-output
stack: database
output: DatabaseEndpointReference outputs from stacks in different AWS regions:
parameters:
SharedBucketArn:
type: stack-output
stack: shared-resources
output: BucketArn
region: us-east-1Support for CloudFormation List<Type> and CommaDelimitedList parameters with mixed resolution types:
parameters:
# Mix literals and stack outputs in a single list parameter
SecurityGroupIds:
- sg-baseline123 # Literal value
- type: stack-output # Dynamic from stack output
stack: security-stack
output: WebSGId
- sg-additional456 # Another literal
# Simple literal list
AllowedPorts:
- "80"
- "443"
- "8080"Different parameter values per deployment context:
parameters:
InstanceType: t3.micro # Default value
contexts:
production:
parameters:
InstanceType: t3.large # Production overrideStack outputs are resolved at deployment time, so cross-stack dependencies always reflect the current live state. Different values per context are supported without modifying templates, and existing literal parameter configurations continue to work unchanged.
Stackaroo supports dynamic CloudFormation template generation using Go templates with Sprig functions. This allows you to use the same template file across different contexts with context-specific variations:
# Template: templates/storage.yaml
AWSTemplateFormatVersion: '2010-09-09'
Description: {{ .Context | title }} storage for {{ .StackName }}
Resources:
AppBucket:
Type: AWS::S3::Bucket
Properties:
BucketName: {{ .StackName }}-bucket-{{ .Context | lower }}
Tags:
- Key: Environment
Value: {{ .Context }}
{{- if eq .Context "production" }}
- Key: BackupEnabled
Value: "true"
{{- end }}When deployed to the development context, this generates a bucket named my-app-bucket-development. When deployed to production, it generates my-app-bucket-production with an additional backup tag.
Templates have access to {{ .Context }} and {{ .StackName }} variables, the full Sprig function library, and standard Go template conditionals for environment-specific resources. Processing is automatic and backwards compatible with static templates.
- Streams live CloudFormation events during deployments, showing resource creation, updates, and completion status as they happen.
- Automatically detects create vs update operations and handles "no changes" scenarios gracefully.
go install codeberg.org/orien/stackaroo@latestDownload the latest release from the releases page.
# Download and install (replace VERSION and ARCH as needed)
VERSION=1.0.0
ARCH=linux-x86_64
URL="https://codeberg.org/orien/stackaroo/releases/download/v${VERSION}/stackaroo-${VERSION}-${ARCH}.tar.gz"
DIR="stackaroo-${VERSION}-${ARCH}"
curl -sL "$URL" | tar -xz
sudo mv "${DIR}/stackaroo" /usr/local/bin/
rm -rf "${DIR}"
# Verify installation
stackaroo --versionDownload the .zip file from the releases page, extract it, and add the binary to your PATH.
stackaroo --versionCreate a stackaroo.yaml file defining your stacks and contexts:
project: my-infrastructure
region: us-east-1
contexts:
development:
account: "123456789012"
region: ap-southeast-4
tags:
Environment: development
production:
account: "987654321098"
region: us-east-1
tags:
Environment: production
stacks:
vpc:
template: templates/vpc.yaml
parameters:
# Literal parameters
Environment: development
VpcCidr: "10.0.0.0/16"
EnableDnsSupport: "true"
contexts:
production:
parameters:
Environment: production
VpcCidr: "172.16.0.0/16"
app:
template: templates/app.yaml
parameters:
# Literal parameters
InstanceType: t3.micro
MinCapacity: "1"
MaxCapacity: "3"
# Stack output parameters (pull from existing stacks)
VpcId:
type: stack-output
stack: vpc
output: VpcId
PrivateSubnetId:
type: stack-output
stack: vpc
output: PrivateSubnet1Id
# Cross-region stack output (optional region parameter)
SharedBucketArn:
type: stack-output
stack: shared-resources
output: BucketArn
region: us-east-1
contexts:
production:
parameters:
InstanceType: t3.small
MinCapacity: "2"
MaxCapacity: "10"
depends_on:
- vpcDeploy stacks using either pattern:
# Deploy all stacks in a context (with dependency ordering)
stackaroo deploy development
# Deploy a specific stack
stackaroo deploy development vpc
# Preview changes before deployment
stackaroo diff development app
# View detailed stack information
stackaroo describe production vpcdeploy <context> [stack-name]- Deploy all stacks or a specific stack with dependency-aware ordering and integrated change previewdiff <context> <stack-name>- Preview changes between deployed stack and local configurationdescribe <context> <stack-name>- Display detailed information about a deployed CloudFormation stackvalidate <context> [stack-name]- Validate CloudFormation templates for syntax and AWS-specific requirementsdelete <context> [stack-name]- Delete stacks with dependency-aware ordering and confirmation prompts
--config, -c- Specify config file (default: stackaroo.yaml)--verbose, -v- Enable verbose output for detailed logging--version- Show version information--help- Show help for any command
# Deploy all stacks in development context
stackaroo deploy development
# Deploy specific stack with verbose output
stackaroo deploy production app --verbose
# Preview changes before deployment
stackaroo diff staging vpc
# View detailed stack information
stackaroo describe production app
# Validate templates before deployment
stackaroo validate development vpc
# Validate all templates in a context
stackaroo validate production
# Delete specific stack with confirmation
stackaroo delete development app
# Delete all stacks in context (reverse dependency order)
stackaroo delete development
# Use custom config file
stackaroo deploy production --config custom-config.yaml