Blocks destructive shell commands and restricts file edits to a chosen directory.
npx skills add https://github.com/abdulazeezoj/monovella-poc --skill guard-abdulazeezoj
Please help me install this Agent Skill. Skill: guard Source: https://github.com/abdulazeezoj/monovella-poc/tree/main/.agents/skills/gstack/guard Command: npx skills add https://github.com/abdulazeezoj/monovella-poc --skill guard-abdulazeezoj
High-intent search queries and answers about installing and using this skill.
Activate guard mode, which registers a PreToolUse hook on Bash that checks every command against destructive patterns such as rm -rf, DROP TABLE, and force-push. Dangerous commands trigger a warning, and catastrophic shapes like recursive deletion of / or ~ are hard-denied.
Guard mode asks which directory to restrict edits to, resolves it to an absolute path, and stores it in a freeze state file. PreToolUse hooks on Edit and Write then block any modification outside that boundary until you run /unfreeze.
No. Guard references hook scripts from the sibling /careful and /freeze skill directories, so both must be installed. They are installed together by the gstack setup script.
Warnings for commands like rm -rf or force-push are overridable after review. However, catastrophic shapes such as recursive deletion of the root or home directory and force-push to the default branch are hard-denied and cannot be overridden.
Run /unfreeze to remove the directory edit boundary. To deactivate all protections entirely, end the session, since the destructive command guard is session-scoped.
256javyEmit warnings before destructive commands and enforce directory edit boundaries.
nocnocgroupClear the freeze state file to lift edit restrictions across directories.
abpaiReview uncommitted Git changes and generate an atomic commit plan.
alexandervashurinEnforce directory-scoped edits and destructive-command warnings during sensitive operations.
