The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes Terraform files (main.tf, README.md, terraform.tfvars) and documentation from external registries, which represents a surface for indirect prompt injection if those files were to contain malicious instructions.
Ingestion points: Template source files and registry metadata are read into the agent's context during the pull, edit, and push workflows.
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore embedded instructions within the ingested code or documentation.
Capability inventory: The skill has access to shell execution via the coder, terraform, gh, and bun CLI tools.
Sanitization: The workflow relies on standard terraform validate and terraform fmt for code integrity but does not mention specific prompt-level sanitization.
[EXTERNAL_DOWNLOADS]: The skill fetches template starters, modules, and documentation from official vendor-controlled domains such as registry.coder.com and coder.com. These are considered safe resources for the skill's intended purpose.
[COMMAND_EXECUTION]: The skill utilizes several command-line tools including coder, terraform, gh, and bun to perform template lifecycle management. These commands are executed within the expected scope of authoring and deploying infrastructure-as-code templates.

templates