Usage (Terraform)

CostCutter ships as a Python package and does not include a Terraform module. You can still orchestrate cleanups from Terraform by invoking the CLI from Terraform resources.

Current Status

• No managed Terraform module or Lambda deployment is bundled with this repository
• All automation relies on the same CLI options documented in CostCutter CLI

Triggering the CLI from Terraform

Example null_resource

resource "null_resource" "costcutter_cleanup" {
	provisioner "local-exec" {
		command = "uvx costcutter --no-dry-run --config ${path.module}/costcutter.yaml"
		environment = {
			COSTCUTTER_LOGGING__LEVEL = "INFO"
		}
	}
}

Key points:

• Package installation is your responsibility (for example via uv tool install costcutter or a virtual environment)
• Provide an explicit config file that lives alongside your Terraform code
• Use --dry-run while validating infrastructure changes and switch to --no-dry-run only when you intend to delete resources

Managing Configuration Files

• Commit a baseline YAML, TOML, or JSON file to your Terraform module
• Populate secrets or overrides with Terraform variables and render them using the templatefile function when required
• Keep the file path stable so the CLI can be invoked in plan and apply stages consistently

Monitoring and Logs

• CLI executions write logs to the directory configured in logging.dir
• Rich output appears in the Terraform apply log; capture it for later review if you use automation
• CSV exports are saved to the path configured under reporting.csv.path

Troubleshooting Terraform Automations

• Ensure the environment running Terraform has AWS credentials usable by CostCutter
• Surface CLI failures by checking Terraform provisioner exit codes and logs
• Run the same command locally to reproduce issues, then consult Troubleshooting for deeper diagnostics