platform-engineer

Files (1)

• SKILL.md

SKILL.md (first lines)

# Platform Engineer Skill

## Overview

Design and architect cloud infrastructure that is secure, production-ready, and
compliant with organization standards. 

Generate Terraform and .tf files based on the architecture proposal outlined in `ARCHITECTURE.md`.

---

## Phase 1: Implementation

Activate the terraform-style-guide skill for more complete instructions on general best practices
to write Terraform. Apply the following organization-specific requirements as well.

### 1.1 Generate provider blocks

`get_latest_provider_version` on the Terraform MCP server to
get the latest provider version before generating code.

When possible, use data sources to retrieve information
from AWS about Kubernetes and other components. If the cluster
name is not specified by the user, prompt the user for
more information.

#### AWS provider block

Check AWS credentials are set up in HCP Terraform using a variable set.

1. Use `list_variable_sets` to get the variables available to the project and workspace
2. Check for required AWS credentials in variable sets:
   - `AWS_ACCESS_KEY_ID` (environment variable)
   - `AWS_SECRET_ACCESS_KEY` (environment variable)
   - `region` (Terraform variable)
3. If AWS credentials are not found in any variable set:
   - Prompt user to attach the appropriate variable set to the workspace
   - Provide instructions: "Attach the AWS credentials variable set to your workspace in HCP Terraform"
4. Declare Terraform variables that match the variable sets

```hcl
# Variable set includes a `region` Terraform variable

variable "region" {
  type        = string
  description = "AWS region"
}
```

**NEVER guess on default tags. If you do not know, prompt the user to specify the correct tags.**

#### Kubernetes provider block

```hcl
data "aws_eks_cluster" "cluster" {
  name = var.cluster_name
}

data "aws_eks_cluster_auth" "cluster" {
  name = var.cluster_name
}

provider "kubernetes" {
  host                   = data.aws_eks_cluster.cluster.endpoint
  cluster_ca_certificate = base64decode(data.aws_eks_cluster.cluster.certificate_authority[0].data)
  token                  = data.aws_eks_cluster_auth.cluster.token
}

provider "helm" {
  kubernetes = {
    host                   = data.aws_eks_cluster.cluster.endpoint
    cluster_ca_certificate = base64decode(data.aws_eks_cluster.cluster.certificate_authority[0].data)
    token                  = data.aws_eks_cluster_auth.cluster.token
  }
}
```

### 1.2 Generate backend.tf file

- Ask user for the HCP Terraform organization and project for deployment
- Suggest a workspace name based on current repository name.
- Create `backend.tf` file in the root of the project to reference HCP Terraform.

```hcl
# backend.tf
terraform {
  cloud {
    organization = "alice-publishing"
    workspaces {
      name = "bobs-book-agent"
    }
  }
}
```

### 1.3 File Structure

Organize Terraform code into standard files:
- `terraform.tf` - Required providers and versions
- `backend.tf` - HCP Terraform configuration
- `variables.tf` - Input variables
- `locals.tf` - Local values
- `main.tf` - Primary resources
- `outputs.tf` - Output values

Refer to terraform-style-guide skill for detailed patterns and examples.

---

## Phase 2: Review and Test

Run `terraform init -no-color` to initialize the state.

Run `terraform fmt -recursive` to format the configuration.

Run `terraform validate -no-color` to validate the configuration.
If validate fails, try to fix and re-validate before proceeding
to plan.

From joatmon08/platform-infrastructure-skills, MIT.

More from joatmon08/platform-infrastructure-skills