设置 Git 防护栏
SkillAI & modelsSets up Claude Code hooks to intercept dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when the user wants to block destructive git operations, add git safety hooks, or intercept git push/reset in Claude Code.
Available today. Use it from your connected AI after setup.
No other account needed.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the 设置 Git 防护栏 skill
What this skill tells your AI
The instructions your AI receives, as published by wenwuzhidao/mattpocock-skills-zh in skills/misc/git-guardrails-claude-code/SKILL.md and read by ahel’s review.
设置一个 PreToolUse 钩子,在 Claude 执行危险的 git 命令之前将其拦截并阻止。
哪些命令会被阻止
git push(包括--force在内的所有变体)git reset --hardgit clean -f/git clean -fdgit branch -Dgit checkout ./git restore .
被阻止时,Claude 会看到一条消息,告知它无权访问这些命令。
步骤
1. 询问适用范围
询问用户:只为当前项目安装(.claude/settings.json),还是为所有项目安装(~/.claude/settings.json)?
2. 复制钩子脚本
随附脚本位于:scripts/block-dangerous-git.sh
根据适用范围将其复制到目标位置:
- 项目级:
.claude/hooks/block-dangerous-git.sh - 全局级:
~/.claude/hooks/block-dangerous-git.sh
用 chmod +x 赋予可执行权限。
3. 在设置中添加钩子
添加到相应的设置文件:
项目级(.claude/settings.json):
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/block-dangerous-git.sh"
}
]
}
]
}
}
全局级(~/.claude/settings.json):
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "~/.claude/hooks/block-dangerous-git.sh"
}
]
}
]
}
}
如果设置文件已存在,将该钩子合并进现有的 hooks.PreToolUse 数组——不要覆盖其他设置。
4. 询问是否自定义
询问用户是否想从阻止列表中添加或移除某些模式。相应地编辑复制后的脚本。
5. 验证
运行一个快速测试:
echo '{"tool_input":{"command":"git push origin main"}}' | <path-to-script>
应以退出码 2 退出,并向 stderr 打印一条 BLOCKED 消息。
Signals
- GitHub stars
- 23
- Forks
- 4
- Last commit
- Aug 2026
Advanced
- Item type
- skill
- Key
git-guardrails-claude-code-wenwuzhidao- Source
- github.com/wenwuzhidao/mattpocock-skills-zh