command-guard

What problem does it solve?

This Skill protects Claude Code projects from accidental data loss, broken Git history, and credential leaks by blocking dangerous shell commands before they execute.

Core Features & Use Cases

• Pre-execution command filtering: Intercepts Bash tool calls and stops risky operations like destructive file deletion, force pushes, hard resets, and database drops.
• Security-focused guardrails: Helps teams enforce safer agent behavior when working with repositories, local environments, and production-like data.
• Customizable policies: Lets you extend the blocklist for project-specific rules such as protected branches, sensitive config files, or deployment restrictions.
• Use case: A developer asks Claude Code to clean a directory or update Git history, and the hook blocks the command until the action is explicitly reviewed.

Quick Start

Ask Claude Code to install the command-guard hook in your project so dangerous Bash commands are blocked before they can run.