The Agent Skills Directory

[SAFE]: The skill is a legitimate development utility designed to assist with software refactoring and maintenance. Its functionality is focused on processing local project files and providing structured feedback based on an architectural scoring rubric.
[COMMAND_EXECUTION]: The skill utilizes local shell commands (Bash/PowerShell) and provided scripts (health-check.sh, health-check.ps1) to perform its analysis. These operations, such as counting line occurrences in .edn files using grep or Select-String, are transparent and appropriately scoped to the project's source directory.
[EXTERNAL_DOWNLOADS]: There are no remote code execution patterns, external package dependencies, or network-based downloads detected in the skill's instructions or scripts.
[PROMPT_INJECTION]: The skill instructions do not contain attempts to override safety guidelines or bypass behavioral constraints. The prompts are focused on defining the agent's role as a modular architecture expert.
[DATA_EXFILTRATION]: The skill lacks network capabilities and does not access sensitive system paths (e.g., .ssh, .aws, or .env). Its data processing is limited to non-sensitive project architecture files within the source/ directory.

module-health