The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using node and potentially yarn node to resolve package paths and query installed metadata.
[DYNAMIC_EXECUTION]: The skill utilizes node -e (evaluate) to run complex JavaScript logic embedded within the instructions. This logic dynamically imports and iterates through the exports of the locally installed es-toolkit library at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided requirements ($ARGUMENTS) to influence the search pattern used in its inspection script. Malicious input could theoretically be crafted to manipulate the generated JavaScript execution string, posing a low-severity injection risk.

recommend