guard

To prevent destructive commands in production, you need safety mode to enforce warnings and require confirmation before executing risky operations like rm -rf or DROP TABLE. It validates commands through pre-flight checks before execution.

Yes, you can restrict file edits to a specific directory by enforcing directory-scoped edit boundaries. This blocks any modifications attempted outside your user-defined safe directory during critical deployments.

PreToolUse hooks enforce edit boundaries by intercepting and validating commands before execution. They check operations against safety conditions and freeze boundaries, stopping risky actions before they run.

The best way to add pre-flight safety checks for debugging workflows is to enable guard mode. It leverages sibling safety components to validate safety conditions before any tool use, protecting critical environments during debugging.

Yes, production protection mode works across both development and production workflows. It covers deployments, debugging, and critical edits, enforcing destructive-command warnings and strict edit boundaries in both environments.

Force-push triggers a destructive-command warning because it is classified as a risky operation that can overwrite critical production state. The safety mode alerts and requires confirmation before executing such commands.