Universal Tool for DevOps and Cloud Automation

automation, cli, cloud, devops, hcl2, helm, helmfile, orchestration, terraform, workflow
brew install atmos


atmos Latest Release Slack Community


Cloud Posse

Universal Tool for DevOps and Cloud Automation.

This project is part of our comprehensive "SweetOps" approach towards DevOps.

It's 100% Open Source and licensed under the APACHE2.


atmos is both a library and a command-line tool for provisioning, managing and orchestrating workflows across various toolchains. We use it extensively for automating cloud infrastructure and Kubernetes clusters.

atmos includes workflows for dealing with:

  • Provisioning Terraform components
  • Deploying helm charts to Kubernetes clusters using helmfile
  • Executing helm commands on Kubernetes clusters
  • Provisioning istio on Kubernetes clusters using istio operator and helmfile
  • Executing kubectl commands on Kubernetes clusters
  • Executing AWS SDK commands to orchestrate cloud infrastructure
  • Running AWS CDK constructs to define cloud resources
  • Executing commands for the serverless framework
  • Executing shell commands
  • Combining various commands into workflows to execute many commands sequentially in just one step
  • ... and many more

In essence, it's a tool that orchestrates the other CLI tools in a consistent and self-explaining manner. It's a superset of all other tools and task runners (e.g. make, terragrunt, terraform, aws cli, gcloud, etc) and intended to be used to tie everything together, so you can provide a simple CLI interface for your organization.

Moreover, atmos is not only a command-line interface for managing clouds and clusters. It provides many useful patterns and best practices, such as:

  • Enforces Terraform and helmfile project structure (so everybody knows where things are)
  • Provides separation of configuration and code (so the same code could be easily deployed to different regions, environments and stages)
  • It can be extended to include new features, commands, and workflows
  • The commands have a clean, consistent and easy to understand syntax
  • The CLI can be compiled into a binary and included in other tools and containers for DevOps, cloud automation and CI/CD
  • The CLI code is modular and self-documenting

CLI Structure

The CLI is built with variant2 using HCL syntax.

*.variant files are combined like Terraform files.

See variant docs for more information on writing commands.

The CLI code consists of self-documenting modules (separating the files into modules is done for cleanliness):

  • utils - a collection of utilities to use in other modules
  • shell - shell commands and helpers for the other modules
  • terraform - terraform commands (plan, apply, deploy, destroy, import, etc.)
  • helm - helm commands (e.g. list)
  • helmfile - helmfile commands (diff, apply, deploy, destroy, sync, lint, etc.)
  • kubeconfig - commands to download and manage kubeconfig from EKS clusters
  • istio - commands to manage istio using istio-operator and helmfile
  • workflow - commands to construct and execute cloud automation workflows

Developing Your Own CLI

One way to use this project is by writing your own custom CLI that leverages our atmos.

This is ideal when you have your own workflows that you want to develop in addition to using the ones we've developed for you.

For example, maybe you have your own existing CLI tools (e.g. using terragrunt). In this case, you may want to start by developing your own CLI.


There are a number of ways you can leverage this project:

  1. As a standalone CLI - you can use our CLI without any modification and get immediate gratification
  2. As a library - you can import our variant modules into your own CLI and expand the workflows for your needs
  3. As a Docker image - you can use our Docker image the way you would the cli and run the workflows

Recommended Layout

Our recommended filesystem layout looks like this:

   │   # CLI configuration
   └── cli/
   │   # Centralized components configuration
   ├── stacks/
   │   │
   │   └── $stack.yaml
   │   # Components are broken down by tool
   ├── components/
   │   │
   │   ├── terraform/   # root modules in here
   │   │   ├── vpc/
   │   │   ├── eks/
   │   │   ├── rds/
   │   │   ├── iam/
   │   │   ├── dns/
   │   │   └── sso/
   │   │
   │   └── helmfile/  # helmfiles are organized by chart
   │       ├── cert-manager/helmfile.yaml
   │       └── external-dns/helmfile.yaml
   │   # Makefile for building the CLI
   ├── Makefile
   │   # Docker image for shipping the CLI and all dependencies
   └── Dockerfile (optional)


The example folder contains a complete solution that shows how to:

  • Structure the Terraform and helmfile components
  • Configure the CLI top-level module main.variant
  • Add stack configurations for the Terraform and helmfile components (to provision them to different environments and stages)
  • Create a Dockerfile with commands to build and include the CLI into the container
  • Combine many CLI commands into workflows and run the workflows to provision resources

In the example, we show how to create and provision (using the CLI) the following resources for three different environments/stages:

  • VPCs for dev, staging and prod stages in the us-east-2 region (which we refer to as ue2 environment)
  • EKS clusters in the ue2 environment for dev, staging and prod
  • nginx-ingress helmfile to be deployed on all EKS clusters

CLI Configuration

The CLI top-level module main.variant contains the global settings (options) for the CLI, including the location of the Terraform components, helmfiles, and stack configurations.

It's configured for that particular example project, but can be changed to reflect the desired project structure.

In the example we have the following:

  • The terraform components are in the components/terraform folder - we set that global option in main.variant
  option "terraform-dir" {
    default     = "./components/terraform"
    description = "Terraform components directory"
    type        = string
  • The helmfiles are in the components/helmfile folder - we set that global option in main.variant
  option "helmfile-dir" {
    default     = "./components/helmfile"
    description = "Helmfile components directory"
    type        = string
  • The stack configurations (Terraform and helmfile variables) are in the stacks folder - we set that global option in main.variant
  option "config-dir" {
    default     = "./stacks"
    description = "Stacks config directory"
    type        = string

main.variant also includes the imports statement that imports all the required modules from the atmos repo.

NOTE: For the example, we import all the CLI modules, but they could be included selectively depending on a particular usage.

    imports = [

NOTE: imports statement supports https, http, and ssh protocols.

NOTE: The global options from main.variant are propagated to all the downloaded modules, so they need to be specified only in one place - in the top-level module.

When we build the Docker image, all the modules from the imports statement are downloaded, combined with the top-level module main.variant, and compiled into a binary, which then included in the Docker container.

Centralized Project Configuration

atmos provides separation of configuration and code, allowing you to provision the same code into different regions, environments and stages.

In our example, all the code (Terraform and helmfiles) is in the components folder.

The centralized stack configurations (variables for the Terraform and helmfile components) are in the stacks folder.

In the example, all stack configuration files are broken down by environments and stages and use the predefined format $environment-$stage.yaml.

Environments are abbreviations of AWS regions, e.g. ue2 stands for us-east-2, whereas uw2 would stand for us-west-2.

$environment-globals.yaml is where you define the global values for all stages for a particular environment. The global values get merged with the $environment-$stage.yaml configuration for a specific environment/stage by the CLI.

In the example, we defined a few config files:

  • ue2-dev.yaml - stack configuration (Terraform and helmfile variables) for the environment ue2 and stage dev
  • ue2-staging.yaml - stack configuration (Terraform and helmfile variables) for the environment ue2 and stage staging
  • ue2-prod.yaml - stack configuration (Terraform and helmfile variables) for the environment ue2 and stage prod
  • ue2-globals.yaml - global settings for the environment ue2 (e.g. region, environment)
  • globals.yaml - global settings for the entire solution

NOTE: The stack configuration structure and the file names described above are just an example of how to name and structure the config files. You can choose any file name for a stack. You can also include other configuration files (e.g. globals for the environment, and globals for the entire solution) into a stack config using the import directive.

NOTE: Currently atmos supports 10 levels of imports, e.g. you can import another config into a stack config, which in turn can import yet another config, etc. See stacks for more details.

Stack configuration files have a predefined format:

    - ue2-globals

    stage: dev



        command: "/usr/bin/terraform-0.13"
            workspace_key_prefix: "vpc"
          cidr_block: ""
            workspace_key_prefix: "eks"

          installed: true


It has the following main sections:

  • import - (optional) a list of stack configuration files to import and combine with the current configuration file
  • vars - (optional) a map of variables for all components (Terraform and helmfile) in the stack
  • terraform - (optional) configuration (variables) for all Terraform components
  • helmfile - (optional) configuration (variables) for all helmfile components
  • components - (required) configuration for the Terraform and helmfile components
  • workflows - (optional) workflow definitions for the stack (see Workflows section below for the more detailed description of workflows)

The components section consists of the following:

  • terraform - defines variables, the binary to execute, and the backend for each Terraform component. Terraform component names correspond to the Terraform components in the components folder

  • helmfile - defines variables and the binary to execute for each helmfile component. Helmfile component names correspond to the helmfile components in the helmfile folder

Run the Example

To run the example, execute the following commands in a terminal:

  • cd example
  • make all - it will build the Docker image, build the CLI tool inside the image, and then start the container

Note that the name of the CLI executable is configurable.

In the Dockerfile for the example, we've chosen the name atmos, but it could be any name you want, for example ops, cli, ops-exe, etc. The name of the CLI executable is configured using ARG CLI_NAME=atmos in the Dockerfile.

After the container starts, run atmos help to see the available commands and available flags.

NOTE: We use the Cloud Posse geodesic image as the base image for the container. This is not strictly a requirement, but our base image ships with all the standard tools for cloud automation that we depend on (e.g. terraform, helm, helmfile, etc).

Provision Terraform Component

To provision a Terraform component using the atmos CLI, run the following commands in the container shell:

  atmos terraform plan eks --stack=ue2-dev
  atmos terraform apply eks --stack=ue2-dev


  • eks is the Terraform component to provision (from the components/terraform folder)
  • --stack=ue2-dev is the stack to provision the component into

Short versions of the command-line arguments can be used:

  atmos terraform plan eks -s ue2-dev
  atmos terraform apply eks -s ue2-dev

To execute plan and apply in one step, use terrafrom deploy command:

  atmos terraform deploy eks -s ue2-dev

Provision Helmfile Component

To provision a helmfile component using the atmos CLI, run the following commands in the container shell:

  atmos helmfile diff nginx-ingress --stack=ue2-dev
  atmos helmfile apply nginx-ingress --stack=ue2-dev


  • nginx-ingress is the helmfile component to provision (from the components/helmfile folder)
  • --stack=ue2-dev is the stack to provision the component into

Short versions of the command-line arguments can be used:

  atmos helmfile diff nginx-ingress -s ue2-dev
  atmos helmfile apply nginx-ingress -s ue2-dev

To execute diff and apply in one step, use helmfile deploy command:

  atmos helmfile deploy nginx-ingress -s ue2-dev


Workflows are a way of combining multiple commands into one executable unit of work.

In the CLI, workflows can be defined using two different methods:

In the first case, we define workflows in the configuration file for the stack (which we specify on the command line). To execute the workflows from workflows in ue2-dev.yaml, run the following commands:

  atmos workflow deploy-all -s ue2-dev

Note that workflows defined in the stack config files can be executed only for the particular stack (environment and stage). It's not possible to provision resources for multiple stacks this way.

In the second case (defining workflows in a separate file), a single workflow can be created to provision resources into different stacks. The stacks for the workflow steps can be specified in the workflow config.

For example, to run terraform plan and helmfile diff on all terraform and helmfile components in the example, execute the following command:

  atmos workflow plan-all -f workflows

where the command-line option -f (--file for long version) instructs the atmos CLI to look for the plan-all workflow in the file workflows.

As we can see, in multi-environment workflows, each workflow job specifies the stack it's operating on:

    description: Run 'terraform plan' and 'helmfile diff' on all components for all stacks
      - job: terraform plan vpc
        stack: ue2-dev
      - job: terraform plan eks
        stack: ue2-dev
      - job: helmfile diff nginx-ingress
        stack: ue2-dev
      - job: terraform plan vpc
        stack: ue2-staging
      - job: terraform plan eks
        stack: ue2-staging

You can also define a workflow in a separate file without specifying the stack in the workflow's job config. In this case, the stack needs to be provided on the command line.

For example, to run the deploy-all workflow from the workflows file for the ue2-dev stack, execute the following command:

  atmos workflow deploy-all -f workflows -s ue2-dev

Share the Love

Like this project? Please give it a ★ on our GitHub! (it helps us a lot)

Are you using this project or any of our other projects? Consider leaving a testimonial. =)

Related Projects

Check out these related projects.

  • variant2 - Turn your bash scripts into a modern, single-executable CLI app today


Got a question? We got answers.

File a GitHub issue, send us an email or join our Slack Community.

README Commercial Support

DevOps Accelerator for Startups

We are a DevOps Accelerator. We'll help you build your cloud infrastructure from the ground up so you can own it. Then we'll show you how to operate it and stick around for as long as you need us.

Learn More

Work directly with our team of DevOps experts via email, slack, and video conferencing.

We deliver 10x the value for a fraction of the cost of a full-time engineer. Our track record is not even funny. If you want things done right and you need it done FAST, then we're your best bet.

  • Reference Architecture. You'll get everything you need from the ground up built using 100% infrastructure as code.
  • Release Engineering. You'll have end-to-end CI/CD with unlimited staging environments.
  • Site Reliability Engineering. You'll have total visibility into your apps and microservices.
  • Security Baseline. You'll have built-in governance with accountability and audit logs for all changes.
  • GitOps. You'll be able to operate your infrastructure via Pull Requests.
  • Training. You'll receive hands-on training so your team can operate what we build.
  • Questions. You'll have a direct line of communication between our teams via a Shared Slack channel.
  • Troubleshooting. You'll get help to triage when things aren't working.
  • Code Reviews. You'll receive constructive feedback on Pull Requests.
  • Bug Fixes. We'll rapidly work with you to fix any bugs in our projects.

Slack Community

Join our Open Source Community on Slack. It's FREE for everyone! Our "SweetOps" community is where you get to talk with others who share a similar vision for how to rollout and manage infrastructure. This is the best place to talk shop, ask questions, solicit feedback, and work together as a community to build totally sweet infrastructure.

Discourse Forums

Participate in our Discourse Forums. Here you'll find answers to commonly asked questions. Most questions will be related to the enormous number of projects we support on our GitHub. Come here to collaborate on answers, find solutions, and get ideas about the products and services we value. It only takes a minute to get started! Just sign in with SSO using your GitHub account.


Sign up for our newsletter that covers everything on our technology radar. Receive updates on what we're up to on GitHub as well as awesome new projects we discover.

Office Hours

Join us every Wednesday via Zoom for our weekly "Lunch & Learn" sessions. It's FREE for everyone!



Bug Reports & Feature Requests

Please use the issue tracker to report any bugs or file feature requests.


If you are interested in being a contributor and want to get involved in developing this project or help out with our other projects, we would love to hear from you! Shoot us an email.

In general, PRs are welcome. We follow the typical "fork-and-pull" Git workflow.

  1. Fork the repo on GitHub
  2. Clone the project to your own machine
  3. Commit changes to your own branch
  4. Push your work back up to your fork
  5. Submit a Pull Request so that we can review your changes

NOTE: Be sure to merge the latest changes from "upstream" before making a pull request!


Copyright © 2017-2021 Cloud Posse, LLC



See LICENSE for full details.

Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements.  See the NOTICE file
distributed with this work for additional information
regarding copyright ownership.  The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License.  You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
KIND, either express or implied.  See the License for the
specific language governing permissions and limitations
under the License.


All other trademarks referenced herein are the property of their respective owners.


This project is maintained and funded by Cloud Posse, LLC. Like it? Please let us know by leaving a testimonial!

Cloud Posse

We're a DevOps Professional Services company based in Los Angeles, CA. We ❤️ Open Source Software.

We offer paid support on all of our projects.

Check out our other projects, follow us on twitter, apply for a job, or hire us to help with your cloud strategy and implementation.


Erik Osterman
Erik Osterman
Andriy Knysh
Andriy Knysh

README Footer Beacon