The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its automation tools. Ingestion points: The scripts pipeline_generator.py, terraform_scaffolder.py, and deployment_manager.py accept user-controlled paths (<project-path>, <target-path>) as input to analyze and generate configuration files. Boundary markers: No explicit delimiters or instructions to ignore embedded instructions within processed project files are defined. Capability inventory: The skill environment facilitates execution of Python, Helm, Terraform, and Kubernetes commands. Sanitization: There is no evidence of input validation or content sanitization for the files being analyzed.
[EXTERNAL_DOWNLOADS]: The install_command defined in metadata.json fetches the skill's documentation from an external, non-whitelisted GitHub repository (borghei/Claude-Skills).
[COMMAND_EXECUTION]: The skill's documentation explicitly instructs the agent to execute three local Python scripts (pipeline_generator.py, terraform_scaffolder.py, and deployment_manager.py) to perform automation tasks.
[NO_CODE]: Although the skill metadata specifies has_scripts: true and describes the functionality of several automation scripts, no executable code files (.py, .sh, etc.) are included in the provided skill package, rendering the core automation features non-functional as distributed.

senior-devops