The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and implement Terraform code, which creates a surface for indirect prompt injection if the agent processes untrusted project files.
Ingestion points: The skill reads local infrastructure requirements, Terraform configuration files (main.tf, variables.tf, outputs.tf), and external cloud environment data via terraform refresh and terraform plan operations in SKILL.md.
Boundary markers: Absent. The instructions do not specify delimiters or provide guidance to the agent to disregard instructions that might be embedded in HCL comments or variable values.
Capability inventory: The skill invokes several shell-based tools including terraform fmt, terraform validate, tflint, terraform plan, and terraform apply across multiple files.
Sanitization: Absent. The agent is directed to address validation errors and state drift based on the raw output of the Terraform CLI tools without a filtering layer.
[EXTERNAL_DOWNLOADS]: The skill references documentation and public repositories for infrastructure tools.
Documentation: Links to documentation hosted at jeffallan.github.io/claude-skills/ are included in SKILL.md for extended guidance.
Community Tools: references/testing.md mentions the use of pre-commit hooks from github.com/antonbabenko/pre-commit-terraform, a widely used community project for Terraform automation.

terraform-engineer