Start Using Workflows

Get Started with CTO.ai Workflows

This page is a technical reference for Workflows usage. For a more conceptual introduction to CTO.ai Workflows, see our Automate with Workflows page.

Our Workflows—in the form of Commands, Pipelines, and Services—can be used to build a cloud application platform that meets your needs.

Workflow SDK Runtimes

Our SDKs provide a straightforward way to run automated processes, send Lifecycle Event data to your CTO.ai Dashboard, and integrate with other systems. We currently offer Workflow SDKs with support for four different runtimes: Python, Node, Bash, and Go.

Anatomy of a Workflow

There are three main components to workflows of every kind:

ops.yml configuration file
Dockerfile (and a corresponding .dockerignore file)
Your custom code

Our ops CLI provides the ops init subcommand to initialize a new Workflow and its corresponding scaffolding files.

> ops init --help
Create a new Workflow

USAGE
  $ ops init [NAME]

ARGUMENTS
  NAME  provide a name or pass a github url to a template

OPTIONS
  -h, --help               show CLI help
  -j, --jobs               generate local template files for pipeline jobs
  -k, --kind=kind          the kind of Application to create (command, pipeline, etc.)
  -t, --template=template  the name of the template to use

Workflow Scaffolding

When you create a Workflow using the ops CLI, it also adds the basic files you might expect when building an application in the language you selected. Here is how all of these files work together, using a Bash-based Workflow as an example:

The main.sh file contains minimal demo code which prompts the user for input and sends a Lifecycle Event to the CTO.ai platform, both using our SDK.
The included Dockerfile defines your automation’s dependencies and runtime environment.
The ops.yml file is used to configure how the Workflow is actually executed. Each Workflow defined in an ops.yml file specifies the container to build and the steps it should run.

Each of our templates includes the basic scaffolding you need to start building a Workflow in the language of your choice. The files you can expect for each language’s Workflow template are listed in the table below:

Common Scaffold Files	Bash	Python	JavaScript	Go
`ops.yml`	✅	✅	✅	✅
`Dockerfile`	✅	✅	✅	✅
`.dockerignore`	✅	✅	✅	✅
Main code file	`main.sh`	`main.py`	`index.js`	`main.go`
Dependencies files	-	`requirements.txt`	`package.json`	`go.mod`

Scaffolding for Pipelines and Jobs

When you initialize a Pipeline, only an ops.yml file will be created.

To generate scaffolding for your Pipeline Jobs, after you have defined one or more Jobs in your ops.yml file, you can run ops init . -j in the directory where your ops.yml file is located. A subdirectory for each Job will be created in ./.ops/jobs/, with the name of each Job being the name of each subdirectory.

Create a Workflow from a Template (ops init)

The CTO.ai ops CLI provides a subcommand to create scaffolding files for your workflow: ops init.

ops init

If you run ops init without any arguments, you will be interactively prompted for a name, the type of workflow you wish to create, a description, and a version number. This information will be used to generate scaffolding code—containing a basic template for your workflow—in a subdirectory of your current path:

A screenshot of a terminal showing the successful output of running `ops init` to generate a Commands workflow.

Build and Run Workflows