The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-provided Terraform requirements and existing infrastructure code as part of its core workflow. This creates a surface for indirect prompt injection where malicious instructions could be embedded in the code comments or metadata of files being reviewed by the agent.
Ingestion points: Reads local Terraform files (.tf, .tfvars) and requirements documentation.
Boundary markers: None explicitly defined in the instructions.
Capability inventory: Executes terraform operations (plan, apply, import, state management) across AWS, Azure, and GCP providers.
Sanitization: No validation or sanitization of external HCL content before processing.
[COMMAND_EXECUTION]: The skill instructs the agent to execute a variety of shell commands and specialized CLI tools including terraform, tflint, opa, conftest, and pre-commit for the purpose of infrastructure automation and testing.
[EXTERNAL_DOWNLOADS]: Fetches and uses various external dependencies from well-known technology providers and official repositories:
Downloads Terraform providers from the official HashiCorp registry.
References Go-based testing frameworks like terratest and testify from Gruntwork's GitHub repository.
Utilizes GitHub Actions for CI/CD workflows from HashiCorp, Bridgecrew, and the Terraform Linters organization.
References official Helm charts for Kubernetes deployments.

terraform-engineer