Infrastructure as Code

Cloud resources defined in reviewed, version-controlled code: a safe state backend, plan review on every change, drift checks, tested modules, policy guardrails and environments you can rebuild from scratch.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Then open the Marketplace in the app, search for Infrastructure as Code, and click Install. It goes into your own files, and works offline from then on.

What it installs.

1

Pillar

1

Area

50

Projects

200

Starter tasks

⚙️ Software Engineering

Infrastructure as Code

Defining servers, networks and cloud resources in version-controlled code so environments are reproducible, reviewable and easy to change.

  1. Inventory of hand-built cloud resources
  2. Choosing an infrastructure as code tool
  3. Remote state backend with locking and encryption
  4. Repository layout for infrastructure code
  5. Pinned tool and provider versions
  6. Environment separation for dev, staging and production
  7. Short-lived credentials for plans and applies
  8. Naming and tagging conventions enforced in code
  9. First production resource brought under code
  10. Infrastructure change rules for the team
  11. Plan review on every infrastructure pull request
  12. Automated plan and apply pipeline for infrastructure
  13. Weekly drift detection run
  14. Monthly provider and module upgrade window
  15. State file health and backup checks
  16. Sweep for resources living outside code
  17. Static checks for infrastructure code
  18. Follow-up on failed applies and stale locks
  19. Secrets kept out of infrastructure code and state
  20. Sandbox account for learning Terraform or OpenTofu
  21. Reading a plan output with confidence
  22. Writing reusable infrastructure modules
  23. State surgery practice in a sandbox
  24. Testing infrastructure code
  25. Server configuration with Ansible
  26. Machine images built with Packer
  27. Team skills matrix for infrastructure code
  28. Importing hand-built production resources into code
  29. Splitting a monolithic state into smaller stacks
  30. Refactoring with moved blocks instead of recreation
  31. Policy as code guardrails in the pipeline
  32. Choosing a runner for infrastructure applies
  33. Evaluating a move from Terraform to OpenTofu
  34. Private module registry with versioned releases
  35. Deletion protection on stateful resources
  36. Environment rebuild from code drill
  37. Major provider version upgrade
  38. Infrastructure code evidence for a compliance audit
  39. Infrastructure code day for the whole team
  40. Solo engineer running a startup's infrastructure
  41. Taking over infrastructure code from a departing engineer
  42. Shared module support for many product teams
  43. Separation of duties for regulated infrastructure changes
  44. Emergency infrastructure change runbook for on-call
  45. Handing infrastructure code to a client
  46. Landing zone of accounts defined in code
  47. GitOps reconciliation for cloud resources
  48. Ephemeral preview environments from infrastructure code
  49. Custom provider for an internal platform
  50. Supply chain checks for modules and providers

Where it starts.

Each project arrives with its purpose, milestones and first tasks written out. Here is how the first three read.

Inventory of hand-built cloud resources

Before writing a line of infrastructure code, you need to know what already exists and who made it. Most teams find resources created in the console years ago, test buckets nobody remembers and security groups with no owner. A single inventory grouped by account, region and service tells you what to import first, what to delete and where the risk sits.

First tasks

  • List every cloud account, subscription or project the team can sign in to
  • Export a resource inventory for each account and active region
  • Mark each resource as managed by code, hand-built or unknown
  • Name an owner for every production resource and flag the ten riskiest

Choosing an infrastructure as code tool

Terraform, OpenTofu, Pulumi, the provider's native templates and the cloud development kits all work, but they differ in licence, language, state handling and how much of your stack they cover. Picking one deliberately, against the clouds and services you actually run and the languages your team already writes, avoids a costly migration two years in.

First tasks

  • Write the list of clouds, SaaS providers and languages the tool must cover
  • Check each candidate's licence terms and provider coverage for those services
  • Build a proof of concept that creates and destroys one storage bucket
  • Record the decision and the runner-up in the team's decision log

Remote state backend with locking and encryption

State on a laptop is one lost disk away from an environment nobody can safely change, and two people applying at once can corrupt it. A remote backend with locking, encryption at rest, versioning and tight access turns state into shared infrastructure rather than a personal file. Set it up before the second engineer touches the code.

First tasks

  • Create the state storage with versioning and encryption in a bootstrap stack
  • Configure locking and test it with two simultaneous plans
  • Restrict access to the pipeline role and named engineers
  • Migrate local state to the backend and add state files to the ignore list

Make it yours.

Install Infrastructure as Code, then tell Aurum the date you're aiming for. It checks whether the plan fits the weeks you actually have.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Blueprints are general information to help you organise your own work. They are not medical, financial, legal or other professional advice. See the Terms.