guard

A safety guard prevents destructive commands like rm -rf or DROP TABLE by intercepting risky operations and issuing warnings before execution. This Skill combines command warnings with directory-scoped edit restrictions to block dangerous operations in live production systems.

Yes, you can restrict file edits to a specific directory path. This Skill enforces directory boundaries by prompting you for a target path, storing the boundary state, and blocking any edit attempts that occur outside the specified directory.

Directory-scoped edit enforcement is a boundary control mechanism that limits file modifications to a user-specified path. It protects production systems by blocking accidental edits outside the designated directory, keeping maintenance and deployment tasks safely isolated.

Yes, this safety guard works for incident response on live deployments. It is designed for production environments where boundary controls are critical, applying destructive-command warnings and edit restrictions during maintenance, deployments, and incident response.

The best way to set up dual safety guards is to compose them from sibling utilities. This Skill implements dual guards by integrating the /careful utility for destructive-command warnings and the /freeze utility for directory-boundary enforcement, guided by an interactive setup prompt.

You need boundary controls during production deployments to prevent catastrophic data loss and system instability. Without dual guards for destructive commands and edit restrictions, accidental force-pushes or unauthorized file modifications can easily disrupt live systems.