guard
Guard mode combines two safety layers: destructive command warnings (rm -rf, DROP TABLE, force-push) and directory-scoped edit restrictions.
Setup & Installation
What This Skill Does
Guard mode combines two safety layers: destructive command warnings (rm -rf, DROP TABLE, force-push) and directory-scoped edit restrictions. It activates both protections at once with a single command. Designed for working on production systems or live debugging sessions where mistakes are hard to reverse.
Instead of manually tracking which files you've touched or second-guessing destructive commands, guard mode enforces both constraints automatically so you can't accidentally overwrite the wrong directory or run a command that deletes data.
When to use it
- Restricting edits to a single service directory while debugging a live API
- Blocking accidental DROP TABLE or rm -rf during a database migration
- Working in a shared repo where only one subdirectory should be touched
- Locking down file writes to a feature branch directory before a production deploy
- Adding a safety net when pair-debugging on a system you don't fully know yet
Similar Skills
best-practices
A checklist of modern web development standards covering HTTPS, CSP headers, input sanitization, deprecated API avoidance, and HTML validity.
auth0-android
Adds authentication to native Android apps using the Auth0 SDK.
auth0-angular
Adds authentication to Angular apps using the @auth0/auth0-angular SDK.
auth0-aspnetcore-api
Adds JWT access token validation to ASP.