The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation extensively utilizes the shell execution syntax (prefixing commands with !) within SKILL.md and various reference files. This is used exclusively for educational purposes to demonstrate how to integrate dynamic bash context into slash commands. Examples include gathering git status or running build scripts.
[COMMAND_EXECUTION]: Static analysis identified 'destructive' system commands such as dd and rm in references/testing-strategies.md. Analysis confirms these are used in the context of teaching users how to perform edge-case testing (e.g., creating a large file with dd to test handling) and standard cleanup procedures, and do not represent a threat to the execution environment.
[PROMPT_INJECTION]: The skill references techniques for input validation and handling user-provided arguments ($1, $ARGUMENTS). While this introduces an indirect prompt injection surface (Category 8), the skill specifically documents 'Validation Patterns' using grep and scripts to mitigate risks associated with processing untrusted user input.

command-development