Warn on destructive commands and restrict edits to a directory.
npx skills add https://github.com/Kit4Some/Oh-my-ClaudeClaw --skill guard-kit4some
Please help me install this Agent Skill. Skill: guard Source: https://github.com/Kit4Some/Oh-my-ClaudeClaw/tree/main/skills/guard Command: npx skills add https://github.com/Kit4Some/Oh-my-ClaudeClaw --skill guard-kit4some
Guard mode unifies destructive-command warnings and edit-scope restrictions to prevent accidental damage during AI-assisted sessions.
Invoke /guard to enable both safety layers for the current session.
High-intent search queries and answers about installing and using this skill.
You can restrict AI edits and warn against destructive commands by enabling a combined guard mode that applies pre-check hooks and edit-scope restrictions to confine changes to a specific directory.
Preventing accidental damage during AI-assisted sessions requires activating safety layers that warn before executing dangerous commands and constrain edits to an allowed directory using state management and pre-check hooks.
To activate both careful warnings and freeze restrictions simultaneously, invoke a single guard command that enables destructive-command warnings and edit-scope limitations with optional directory targeting for the current session.
Edit-scope freezing works by applying pre-check hooks that intercept actions, logging them for traceability, and blocking any modifications outside the specifically targeted directory during the AI session.
Yes, when enabling edit restrictions you can optionally target a specific directory, ensuring all AI-assisted modifications are confined exclusively to that path while logging actions for traceability.
Intension-usWarn on destructive commands and block edits outside a chosen directory.
ScivorRemove the /freeze border to allow edits across all directories.
wuyoscarEvaluate model responses as safe or unsafe using Llama-Guard templates.
TioSavichRestrict edits to a specified directory and warn before destructive commands.
