菜单
Install a Claude Code hook that stops destructive git commands (push, reset --hard, clean, branch -D, and friends) before they run. Use when the user wants to guard against dangerous git operations, add git safety hooks, or block git push/reset inside Claude Code.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Recurring maintenance pass over automated dependency-bump PRs (Dependabot, Renovate, or similar) — rebase, relock, verify, and report what's ready to merge. Meant to be handed to a time-based loop or schedule, not run once.
Post-deploy verification loop — poll a rollout until every instance is on the new version and healthy, then run one smoke check against a real user path. Meant to be handed to a time-based loop with a timeout, not polled by hand.
Reference for designing agent loops — cycles of work that repeat until a stop condition is met. Covers the four loop shapes, writing completion criteria, carrying state and isolating work across cycles, and what running unattended still leaves on you.
Recurring check-in on a stack of dependent PRs — rebase children onto updated parents, surface CI state, flag anything waiting on a human-only gate, and report only what changed since last time. Meant to be handed to a time-based loop, not run once.
Pick the skill or flow that fits your situation. A router over the skills in this repo.
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/PRD asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".
| name | git-guardrails-claude-code |
| description | Install a Claude Code hook that stops destructive git commands (push, reset --hard, clean, branch -D, and friends) before they run. Use when the user wants to guard against dangerous git operations, add git safety hooks, or block git push/reset inside Claude Code. |
Wire up a PreToolUse hook that inspects every Bash command Claude is about to run and refuses the destructive git operations before they touch the repository.
git push in any form, including --forcegit reset --hardgit clean -f and git clean -fdgit branch -Dgit checkout . and git restore .When a command matches, the hook exits with code 2 and Claude receives a message stating it is not permitted to run that command, so it stops rather than retrying.
Ask whether the guard should cover this project only (.claude/settings.json) or every project (~/.claude/settings.json). The answer decides where the script and the hook entry go.
The bundled script lives at scripts/block-dangerous-git.sh. Copy it to match the chosen scope:
.claude/hooks/block-dangerous-git.sh~/.claude/hooks/block-dangerous-git.shThen mark it executable with chmod +x.
Add the hook to the settings file for the chosen scope.
Project (.claude/settings.json):
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "\"$CLAUDE_PROJECT_DIR\"/.claude/hooks/block-dangerous-git.sh"
}
]
}
]
}
}
Global (~/.claude/settings.json):
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "~/.claude/hooks/block-dangerous-git.sh"
}
]
}
]
}
}
When the settings file already exists, merge this entry into the existing hooks.PreToolUse array and leave the rest of the file untouched.
Ask whether any patterns should be added or dropped, then edit the DANGEROUS_PATTERNS array in the copied script to match.
Feed a sample command through the script and check the result:
echo '{"tool_input":{"command":"git push origin main"}}' | <path-to-script>
A working guard exits with code 2 and prints a BLOCKED line to stderr.