Manages Kubernetes manifests and Terraform infrastructure for project environments.
npx skills add https://github.com/oporpino/commons --skill br4zz4-infra-oporpino
Please help me install this Agent Skill. Skill: br4zz4:infra Source: https://github.com/oporpino/commons/tree/main/ai/shared/skills/br4zz4%3Ainfra Command: npx skills add https://github.com/oporpino/commons --skill br4zz4-infra-oporpino
.infra/k8s/base/specs/ with per-environment overlays patching ingress hosts, replica counts, and image tags set by CI. - Terraform with S3 backend and workspaces: State stored in br4zz4-terraform-states, with staging and production workspaces driving namespaces (<app>--<env>), domains, and resource naming. - Secrets via ward vaults and SOPS: Per-project ward vaults at .ward/, central org vaults in bootstrap and commons repos, and SOPS-encrypted secrets/<env>.enc.yml files loaded by the br4zz4 wrapper. - Use Case: When adding a new project, you scaffold .infra/terraform/ with the two-provider AWS pattern (shared account for EKS auth, project account for resources), bootstrap a TerraformRole, and run make terraform.staging.apply to create the namespace, DNS record, secrets, and workloads from scratch. ## Quick Start Ask the assistant to create the Terraform and Kustomize configuration for a new project's staging environment following the infra conventions.High-intent search queries and answers about installing and using this skill.
Use one Terraform configuration with two workspaces, staging and production, sharing the same code but separate state in the S3 backend. The workspace drives the namespace (`<app>--<env>`), domain, and resource names via `${local.environment}`.
Store secrets in a per-project ward vault at `.ward/`, organized by app, infra level, and environment. CI reads them via `ward exec <path> -- <command>`, and DATABASE_URL is never stored since Terraform generates it from RDS outputs.
The EKS cluster token minted at the start of the run expires during long applies, such as RDS restores taking 6-12 minutes. Re-run the apply or target only the Kubernetes resource with a fresh token, or split slow database work and Kubernetes jobs into separate applies.
No. App resources belong in the project's own account VPC, peered to the shared VPC, and traffic flows over private IPs through the peering connection. Placing app databases in the shared VPC pollutes the platform network and removes environment isolation.
Mixing inline `route {}` blocks in `aws_route_table` with standalone `aws_route` resources causes the table to delete any route not declared inline on every apply. Declare all routes as standalone `aws_route` resources and import existing inline routes before applying.
boundless-xyzAutomate Boundless AWS dev infrastructure deployment and teardown with Pulumi and AWS CLI.
codewithmukeshAutomate .NET build, test, and deployment pipelines with GitHub Actions or Azure DevOps.
MassimilianoPiliGenerate and deploy AWS CloudFormation templates for Amazon RDS instances and clusters.
wzyxdwllAutomate cloud-native infrastructure delivery across Kubernetes, GitOps, and IaC tooling.