The Agent Skills Directory

[SAFE]: The skill performs standard code generation tasks within the scope of its defined purpose.
[COMMAND_EXECUTION]: Uses Grep and Glob to check for existing implementations in the project codebase, which is a common and safe practice for avoiding code conflicts during generation.
[DATA_EXFILTRATION]: No evidence of sensitive data collection or external network operations. The templates use UserDefaults locally on the device to track app session data, which is standard behavior for this functionality.
[PROMPT_INJECTION]: Instructions focus on technical implementation and platform requirements (e.g., App Store distribution) without attempting to bypass agent constraints or safety filters.
[REMOTE_CODE_EXECUTION]: The skill does not download external scripts or execute remote code; it generates static Swift files from local templates.

review-prompt