The Agent Skills Directory

[COMMAND_EXECUTION]: The skill automates Terraform operations such as terraform init and terraform providers schema using shell commands. It also utilizes jq for data processing and curl to fetch provider versions. A local bash script, lib/validate_template.sh, is executed to perform automated linting and validation of the generated templates.
[EXTERNAL_DOWNLOADS]: The skill workflow involves downloading provider information from the official HashiCorp Terraform Registry (registry.terraform.io) and accessing documentation/MCP server configurations via official AWS domains (api.aws). These sources are recognized as well-known and trusted technology services.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied infrastructure descriptions to generate Terraform code, creating a surface for potential injection attacks. 1. Ingestion points: The {概要} parameter within the /terraform-aws-annotated-blueprint command. 2. Boundary markers: Absent; there are no instructions provided to the agent to treat user input as untrusted or to use specific delimiters. 3. Capability inventory: The skill can write files to the local .local directory and execute multiple shell-based commands (terraform, jq, curl, bash). 4. Sanitization: The skill does not define any input validation or sanitization routines for the user-provided summary.

terraform-aws-annotated-blueprint