---
menu:
  learn:
    parent: Patterns quick start
title: Creating a pattern with patternizer
weight: 19
---

:toc:
:imagesdir: /images
:_content-type: ASSEMBLY
include::modules/comm-attributes.adoc[]

== Creating a new pattern with patternizer

link:https://github.com/validatedpatterns/patternizer[Patternizer] is a command-line tool that bootstraps a Git repository into a ready-to-use Validated Pattern. It generates all required scaffolding -- values files, Makefile, `pattern.sh`, and Ansible configuration -- so you can focus on defining your applications rather than wiring up framework boilerplate.

=== Prerequisites

* A container runtime (Podman or Docker)
* Git

[TIP]
.Shell Function
====
You can add a simple shell function to your shell configuration file (for example `~/.zshrc` or `~/.bashrc`) to enable easier use of the tool.

[source,bash]
----
pattern() {
  podman run --pull=newer \
    -v "$PWD:$PWD:z" \
    -w "$PWD" \
    quay.io/validatedpatterns/patternizer "$@"
}
----

With this function, you can run `pattern init` directly instead of the full `podman run` command. The examples below assume this function is in place.
====

=== Step 1: Create your pattern repository

Create or clone an empty Git repository named after your pattern. The directory name becomes the pattern name in `values-global.yaml`.

[source,terminal]
----
$ mkdir my-pattern && cd my-pattern
$ git init
----

=== Step 2: Initialize the pattern

To create a pattern without the secrets framework:

[source,terminal]
----
$ pattern init
----

To create a pattern with HashiCorp Vault and External Secrets Operator scaffolding:

[source,terminal]
----
$ pattern init --with-secrets
----

==== What patternizer generates

Running `pattern init --with-secrets` on an empty directory named `my-pattern` produces the following files:

[source,bash]
----
my-pattern
├── ansible.cfg                 # default Ansible configuration
├── Makefile                    # stub Makefile which includes Makefile-common
├── Makefile-common             # common commands (install, load-secrets, etc.)
├── pattern.sh                  # convenience utility container (has oc, helm, make, etc.)
├── values-global.yaml          # names the pattern based on the directory and sets common defaults
├── values-prod.yaml            # contains the components for the secrets framework
└── values-secret.yaml.template # stub for the secrets file
----

The generated `values-global.yaml` contains:

[source,yaml]
----
global:
  pattern: my-pattern
  singleArgoCD: true
  secretLoader:
    disabled: false
main:
  clusterGroupName: prod
  multiSourceConfig:
    enabled: true
    clusterGroupChartVersion: 0.9.*
----

The generated `values-prod.yaml` contains:

[source,yaml]
----
clusterGroup:
  name: prod
  namespaces:
    my-pattern:
    vault:
    external-secrets-operator:
      operatorGroup: true
      targetNamespaces: []
    external-secrets:
  subscriptions:
    eso:
      name: openshift-external-secrets-operator
      namespace: external-secrets-operator
      channel: stable-v1
  applications:
    openshift-external-secrets:
      name: openshift-external-secrets
      namespace: external-secrets
      chart: openshift-external-secrets
      chartVersion: 0.0.*
    vault:
      name: vault
      namespace: vault
      chart: hashicorp-vault
      chartVersion: 0.1.*
----

If you omit `--with-secrets`, the generated files will not include the Vault and External Secrets Operator configuration, and `secretLoader.disabled` will be set to `true` in `values-global.yaml`.

=== Step 3: Define your pattern content

After initialization, you need to add your operators, applications, and Helm charts to the pattern. There are two approaches:

==== Using the pattern-author AI coding skill

Patternizer installs an AI coding skill to `.claude/skills/pattern-author/` and `.cursor/skills/pattern-author/` during initialization. This skill teaches AI coding assistants such as Claude Code and Cursor how to author Validated Patterns -- defining namespaces, operators, applications, secrets, hub/spoke clusters, and more.

Open the initialized repository in your AI-assisted editor and the skill will be available automatically. You can ask the assistant to add operators, wire in Helm charts, configure secrets, or set up multi-cluster deployments.

==== Manually editing values files

You can directly edit the generated values files to define your pattern:

* Add operator subscriptions and namespaces to `values-prod.yaml`
* Add Helm chart applications to `values-prod.yaml`
* Configure global settings in `values-global.yaml`
* Define secrets in `values-secret.yaml.template`

For detailed guidance on structuring your pattern, see link:/learn/vp_structure_vp_pattern/[Structuring a validated pattern]. For adding operators, see link:/learn/vp_add_ops_to_pattern/[Adding operators to the framework].

[NOTE]
.Idempotency
====
The `pattern init` command is idempotent and can be run multiple times during pattern creation to update the pattern values files. You can go from `pattern init` to `pattern init --with-secrets` to add the secrets framework to your pattern. If you use `helm create` (or `./pattern.sh helm create`) to create Helm charts and then run `pattern init` again, the Helm charts will be automatically added to your `values-prod.yaml`.
====

=== Next steps

* Deploy an existing pattern like link:/learn/getting-started-multi-cloud-gitops/[Multicloud GitOps] to explore how the framework works in practice
* Learn about link:/learn/vp_structure_vp_pattern/[structuring a validated pattern] and best practices
* Add operators to your pattern with link:/learn/vp_add_ops_to_pattern/[Adding operators to the framework]
* Configure secrets with link:/learn/getting-started-secret-management/[Configuring secrets]
* Visit the link:https://github.com/validatedpatterns/patternizer[patternizer repository on GitHub] for the latest documentation