---
title: "claude-code-safety-net"
description: "Hook that intercepts destructive shell commands before an autonomous agent executes them."
canonical: "https://vibing.inc/library/claude-code-safety-net-vg-0016"
category: "AI Skills, Agents & MCP"
subcategory: "Agent safety"
license: "open-source"
author: "@kenryu42"
source: "https://github.com/kenryu42/claude-code-safety-net"
---

# claude-code-safety-net
Hook that intercepts destructive shell commands before an autonomous agent executes them.
## Facts
- Source: https://github.com/kenryu42/claude-code-safety-net
- Category: AI Skills, Agents & MCP / Agent safety
- Author: @kenryu42
- Licence: open-source
- Tags: claude-code, hooks, safety, guardrails
## Why it's a gem
Twenty lines between you and an agent running rm -rf on your repo.
---
Listed in The Vibeary: https://vibing.inc/library/claude-code-safety-net-vg-0016 · JSON: https://vibing.inc/api/public/library/gems/claude-code-safety-net-vg-0016