Refactor SC tagging to ensure consistent tags on all resources (AWS + GCP) and auto-apply on existing resources

[Cre-eD]

Description

We need to refactor how tags (AWS) / labels (GCP) are handled in SC.

Currently, tagging is not consistently applied to all resources, which limits our ability to use cost allocation and Cost Explorer features effectively.

We need a unified and reliable tagging strategy that:

1. Applies required tags to all supported resources during creation.
2. Automatically adds missing tags to existing resources on subsequent runs, without causing errors or breaking changes.
3. Works consistently across AWS (tags) and GCP (labels).

---

Required Tags / Labels

The following metadata must be applied to all applicable resources:

• parent_stack_name
• client_stack_name
• env_name

Naming convention can follow existing SC standards (snake_case or existing convention).

---

Requirements

1. Tagging During Resource Creation

• All SC-managed resources must include the required tags at creation time.
• Applies to all supported AWS resource types.
• Applies to all supported GCP resource types (via labels).

2. Tag Backfill for Existing Resources

• On a new SC run:

  • If a managed resource is missing required tags/labels,
  • SC should add them automatically,
  • Without recreating the resource,
  • Without producing errors or destructive changes.

This should be idempotent and safe.

3. Cross-Cloud Consistency

• AWS → use tags
• GCP → use labels
• Logical structure and naming must remain consistent between clouds.

---

Motivation

• Enable proper use of AWS Cost Explorer and cost allocation reporting.

• Enable similar cost analysis capabilities in GCP.

• Ensure consistent cost visibility by:

  • Parent stack
  • Client stack
  • Environment

• Reduce manual tagging and human error.

• Improve governance and FinOps practices.

---

Acceptance Criteria

• All newly created SC resources contain required tags/labels.
• Running SC on existing stacks automatically applies missing tags.
• No resource recreation is triggered solely due to missing tags.
• Behavior is consistent across AWS and GCP.
• Cost Explorer / billing reports can filter by required tags.

[simple-container-forge]

✅ Workflow Completed Successfully!

Great news! The Simple Forge has successfully processed your request.

📋 Job Details:

• Job ID: bd9567ec-5fd2-4dfc-9793-46f0f5bbb47b
• Job URL: https://forge.simple-container.com/jobs/bd9567ec-5fd2-4dfc-9793-46f0f5bbb47b
• Branch: simple-forge/issue-154-refactor-sc-tagging-to-ensure-tartkz
• Workflow: View Results
• Completed: 2026-02-20 19:08:07 UTC

📝 Execution Summary:
I've processed the issue and made the following changes:

I'll analyze this issue about refactoring SC tagging to ensure consistent tags across AWS and GCP resources. Let me start by exploring the codebase to understand the current state.

Now I have a comprehensive understanding of the codebase. Let me create a todo list and start implementing the solution:

Let me first read the existing files to understand the exact implementation:

Now let me read the GCP files to understand the current implementation:

Now I have a complete understanding of the codebase. Let me create the unified tagging utility and then update all the necessary files. First, I'll create the tagging utility package:

Now let me update the todo list and proceed with updating the AWS files:

Now let me update the AWS ECS Fargate file to use the new tagging utility:

Now let me update the AWS static egress file to use the new tagging utility. First, I need to add the tags to all resources in that file:

Let me read the file again to see the current state:

I see the issue - I changed the variable name to baseTags but there are still references to tags. Let me fix this:

Now I need to update the zoned tags to also include the SC metadata. Let me read more of the file to understand the structure:

I need to update the zonedTags to include the SC metadata tags. Let me do that:

Now I need to update the KMS key tagging:

Now let's update the GCP Artifact Registry to use labels:

Now let me check if there are any GCP compute resources that need labeling. Let me search for GCP compute files:

Let me read the GCP compute_proc.go file:

This file is about Cloud SQL compute processors, not about labeling GCP compute resources. Let me check the GKE Autopilot files instead:

This is just a stub. Let me search for actual GKE cluster creation files:

Let me read the GKE Autopilot stack file:

The GKE Autopilot stack file doesn't directly create GKE resources with labels - it just references existing clusters. Let me now check if there's actual cluster creation code:

This file creates the GKE cluster but doesn't set labels. Let me update it to add labels to the cluster and other GCP resources. First, let me update the todo list and then make the changes:

Now let me update the GKE Autopilot file to add labels to the cluster and VPC resources:

Now let me update the cluster creation to add labels:

Now let's update the Kubernetes resource labeling to ensure consistency. Let me update the simple_container.go file to use consistent labels that match our tagging convention:

Now let me update the Kubernetes labels to be consistent with the tagging utility. The existing labels are already quite good but let me add the missing parent_stack and client_stack labels:

Now let me update the code that creates the appLabels to include these new labels:

Now let's verify all our changes by reading the modified files:

Let me verify the changes by reading key parts of the modified files:

Let me verify the AWS ECS Fargate changes:

Let me verify the Kubernetes labeling changes:

Now let me verify the appLabels section:

Let me also check the GKE autopilot file to verify the cluster labels:

Let me check the cluster args with labels:

All the changes look good. Let me finalize by updating the todo list:

Let me run the build to make sure there are no compilation errors:

The builds are downloading dependencies. Let me check if there are any compilation errors by waiting for the build to complete:

The file was created successfully. Let me check if there are any issues by compiling the main command:

I see the issue - the ProvisionParams struct doesn't have StackName and Environment fields directly. Let me check what fields are available:

I see - the ProvisionParams doesn't have StackName and Environment directly. I need to get these from input.StackParams instead. Let me check the function signature of provisionStaticEgressForMultiZoneVpc:

[Content truncated - full details available in workflow logs]

🎉 Next Steps:

• Review the changes in the branch simple-forge/issue-154-refactor-sc-tagging-to-ensure-tartkz
• Test the implementation
• Create a pull request when ready

Feel free to ask for any modifications or improvements!

[simple-container-forge]

✅ Workflow Completed Successfully!

Great news! The Simple Forge has successfully processed your request.

📋 Job Details:

• Job ID: 7390ff05-dc90-4fc4-9a39-c86dae8ec977
• Job URL: https://forge.simple-container.com/jobs/7390ff05-dc90-4fc4-9a39-c86dae8ec977
• Branch: simple-forge/issue-154-refactor-sc-tagging-to-ensure-tartkz
• Workflow: View Results
• Completed: 2026-02-20 19:30:07 UTC

📝 Execution Summary:
I've processed the issue and made the following changes:

I'll analyze the issue and implement the necessary changes to refactor SC tagging for consistent tags across AWS and GCP resources.

Let me start by exploring the codebase structure to understand the current tagging implementation:

Now let me read key files to understand the current tagging implementation:

Let me explore the API models to understand the StackParams structure:

Now I have a comprehensive understanding of the codebase. Let me create a unified tagging utility and then update all the necessary files. First, I'll create the tagging utility package:

Now let me update the todo list and proceed with updating the AWS files:

Now let me update the AWS ECS Fargate file to use the new tagging utility:

Now let me read the static_egress.go file to update it:

Let me read more of the static_egress.go file:

INow I need to update the static_egress.go file to use the new tagging utility. I'll add the import and update the tags:

Now I need to add the StackParams to the StaticEgressIPIn struct and update the tags. Let me first update the struct:

Now let me update the tags in the provisionStaticEgressForMultiZoneVpc function:

Now I need to update all references to tags to use baseTags in this function:

Now I need to update the zoned tags to also include the SC metadata. Let me read the file again to see the current state:

I need to update the zoned tags to include the SC metadata tags. Let me update this:

Now I need to update the KMS key file. Let me read it first:

Now let me update the KMS key file to use the new tagging utility:

Now let me check the GCP Artifact Registry file:

Now let me update the GCP Artifact Registry to add labels:

Now let me update the todo list and proceed with updating the GKE Autopilot file:

Now let me check if there are any GCP compute resources that need labeling. Let me search for GCP compute files:

Let me check the GKE Autopilot file to see if there's a place to add labels:

Now let me update the GKE Autopilot file to add labels to the cluster:

Now let me add labels to the GKE cluster:

Now let me update the Kubernetes resources to use consistent labels. Let me update the todo list first:

Now let me check the Kubernetes simple_container.go file to see if it needs any label updates. Looking at the file I read earlier, I can see it already has labels like:

• LabelAppType = "appType"
• LabelAppName = "appName"
• LabelScEnv = "appEnv"
• LabelParentEnv = "simplecontainer.com/parent-env"
• LabelCustomStack = "simplecontainer.com/custom-stack"

These labels are already quite good but I should add the missing parent-stack and client-stack labels to be consistent with our tagging utility. Let me update the simple_container.go file:

Now let me update the code that creates the appLabels to include these new labels:

Now let me update the todo list and verify all changes by reading the modified files:

Now let me verify the changes by reading key parts of the modified files:

Let me verify the AWS ECS Fargate changes:

Let me check the function that uses the tags:

Good! Now let me verify the Kubernetes labeling changes:

Let me verify the appLabels section:

Good! Now let me also check the GKE autopilot file to verify the cluster labels:

Excellent! Now let me finalize by updating the todo list and running a quick build check:

Now let me run a quick build check to make sure there are no compilation errors:

I see the issue - the input.StackParams is a pointer, but the function expects a value. Let me fix this:

Now I need to fix the same issue in the static_egress.go file:

Wait, I made a mistake. Let me check the static_egress.go file again to see how StackParams is used:

I see that in static_egress.go, the StackParams is a value, not a pointer. But I'm using ToGCPLabels() which returns a map[string]string, but I need AWS tags. Let me fix this:

[Content truncated - full details available in workflow logs]

🎉 Next Steps:

• Review the changes in the Pull Request #155
• Test the implementation
• Merge the pull request when ready

Feel free to ask for any modifications or improvements!