The Agent Skills Directory

[PROMPT_INJECTION]: No malicious injection patterns detected. The skill uses strong instructional keywords to ensure its own internal logic and safety protocols (like the visual inspection gate) are followed.
[COMMAND_EXECUTION]: Uses Bash for legitimate development tasks like capturing simulator screenshots (xcrun simctl), checking git status, and performing file backups. All commands are standard for an iOS development workflow.
[EXTERNAL_DOWNLOADS]: Recommends the installation of a third-party tool (Dippy) to the user for path-handling convenience, but does not execute external code or perform automated downloads itself.
[DATA_EXFILTRATION]: No evidence of data exfiltration. Network tools are not used by the skill's logic.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or access to sensitive credential files detected.

ui-enhancer-radar