Block destructive commands and file edits outside a specified directory.
npx skills add https://github.com/Shmurdoc/EasyRide-v1 --skill guard-shmurdoc
Please help me install this Agent Skill. Skill: guard Source: https://github.com/Shmurdoc/EasyRide-v1/tree/main/.opencode/skills/gstack-guard Command: npx skills add https://github.com/Shmurdoc/EasyRide-v1 --skill guard-shmurdoc
When working on production systems or debugging live environments, accidental execution of destructive commands (e.g., rm -rf, DROP TABLE, force-push) or unintended edits to files outside your target directory can lead to data loss, service outages, and costly rollbacks.
Activate guard mode by asking the AI to enable full safety mode and provide the directory path you want to restrict all file edits to.
High-intent search queries and answers about installing and using this skill.
To prevent accidental destructive commands during live debugging, you can activate full safety mode to enforce real-time warnings for high-risk command patterns like rm -rf, DROP TABLE, and git reset --hard.
Yes, you can restrict file edits to a specific directory by enabling full safety mode, which locks all file edit operations to your chosen path and blocks modifications to any files outside that boundary.
The best way to add guardrails for production system maintenance is to enable dual safety protections, combining real-time warnings for destructive commands with strict edit boundary enforcement to prevent data loss and service disruption.
To block force-push and git reset --hard from executing accidentally, enable a safety mode that monitors command patterns in real-time and issues immediate warnings before high-risk destructive commands are executed.
Yes, edit scoping works for preventing unintended changes by locking all file edit operations to a user-specified directory, strictly blocking any modifications to critical system files outside that defined boundary.
A limitation of using this safety mode is that it only provides warnings for destructive commands and blocks edits outside a specified directory, meaning it relies on user-specified boundaries to define the scope of protection.
Walrus-ComputingWarn before destructive shell commands and restrict file edits to an approved directory.
balajivisWarn before destructive commands and restrict file edits to a selected directory.
nuurproWarn on destructive commands and restrict edits to a specified directory.
abhishekdutta18Block destructive commands and restrict file edits to specified directories.
