The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the curl tool to send generated Eraser DSL code to the Eraser API endpoint (https://app.eraser.io/api/render/elements) for diagram rendering.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data in the form of Terraform files (.tf, .tfvars) to extract infrastructure relationships. This represents an indirect prompt injection surface where malicious instructions could be embedded in HCL comments or string values within the infrastructure code.
Ingestion points: Reads resource, module, data, and variable blocks from user-provided Terraform files.
Boundary markers: The skill does not explicitly define boundary markers or delimiters for the ingested HCL content.
Capability inventory: The skill has access to the Read tool for file ingestion and Bash (specifically curl) for network transmission to the vendor's API.
Sanitization: While the skill enforces specific formatting rules for the generated Eraser DSL labels, it does not specify sanitization or validation logic for the content extracted from the Terraform files.

terraform-diagrams