What problem does it solve? Manually provisioned cloud infrastructure drifts from intent, cannot be audited, and cannot be reproduced when a new environment is needed. This Skill replaces ClickOps with version-controlled Terraform or Pulumi code that is reviewable, repeatable, and security-scanned before apply. ## Core Features & Use Cases - Structured Terraform scaffolding: Generates versions.tf with pinned providers, backend.tf with remote state and locking (S3+DynamoDB, GCS, or Azure Blob), typed variables.tf, outputs.tf, and per-environment .tfvars files. - CI plan/apply pipeline: Adds GitHub Actions jobs that run tfsec security scans, produce reviewable plans on pull requests, and apply on merge to main with environment-gated production deploys. - Golden-path modules: Supports consuming curated, versioned modules from a private registry so services inherit compliant defaults (encryption, tagging, network policy) instead of hand-writing bespoke HCL. - Use Case: A service needs a new RDS database, S3 bucket, and SQS queue. The Skill generates the full infra/ directory, a staging/production tfvars split, and a CI pipeline, then runs the iac-reviewer agent before committing. ## Quick Start Ask the AI to provision the infrastructure for a new service using Terraform with remote state, a tfsec scan, and a staging/production environment split.
