The Agent Skills Directory

[SAFE]: The skill serves as a documentation and template guide for infrastructure provisioning. It contains no executable malicious scripts and focuses on educational instructions for project structure.
[EXTERNAL_DOWNLOADS]: The instructions reference the official AWS provider from HashiCorp. HashiCorp is a well-known and established service provider in the infrastructure space, and downloads from its official registries are considered safe.
[COMMAND_EXECUTION]: The skill describes the use of standard, non-malicious Terraform commands (terraform fmt, terraform validate, terraform plan, terraform apply) intended for use in local development and standard CI/CD pipelines.
[DATA_EXFILTRATION]: No data exfiltration patterns were detected. On the contrary, the skill explicitly provides security guidance to avoid committing sensitive state files and .tfvars secrets to version control.
[PROMPT_INJECTION]: Analysis of indirect injection surfaces:
Ingestion points: User-provided infrastructure requirements and cloud provider details.
Boundary markers: The skill does not explicitly define markers but operates as a passive template guide.
Capability inventory: Filesystem write operations for Terraform configuration files.
Sanitization: None specified within the skill body, but instructions emphasize safe secret management via environment variables and gitignore patterns.

setting-up-terraform