
Git Guardrails for Claude Code
Blocks risky git commands like push, reset --hard and branch -D before Claude runs them.
- Repository
- mattpocock/skills
- Stars
- 309,658
- Licence
- MIT
- Files
- 3 files
What it does
This skill sets up a safety hook in Claude Code, the coding assistant. Once installed, a small script checks every shell command Claude tries to run and stops git commands that could overwrite or discard work, such as pushing, hard-resetting, cleaning files, deleting branches, or discarding local changes. Claude receives a message saying the action was blocked, so it does not run it.
Great for
- Developers who let Claude Code work in their repositories and want a backstop against destructive git actions
- Teams that want to keep Claude from pushing code directly to shared branches
- People new to AI coding agents who want extra protection while they learn how the tool behaves
Skip it if
- Users who need Claude to push, commit to remote, or reset branches as part of normal work, since those commands will be blocked and must be run manually
- Non-git projects or non-Claude-Code agents, because the hook relies on Claude Code's PreToolUse hook system and a jq-based script
- Anyone expecting complete protection: the pattern matching is simple text matching and can be bypassed by alternate command forms, and it only covers Bash tool calls
Try asking
- “Set up git guardrails so Claude can't push or hard-reset in this project”
- “Add a hook that blocks git branch -D and git clean in all my projects”
- “Block git push and git reset in Claude Code for this repo”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 3 original files, the MIT licence and an attribution note.
git-guardrails-claude-code.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip git-guardrails-claude-code.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip git-guardrails-claude-code.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Choose project scope for shared repos so teammates are not affected by a personal global hook
- Test the hook with the included echo command before relying on it, and confirm it exits with code 2
- Review the pattern list and add or remove entries to fit your workflow, since broad patterns like 'git push' also match harmless text in commands
What's inside
- agents/openai.yaml
- scripts/block-dangerous-git.sh
- SKILL.md
Redistributed unmodified from mattpocock/skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.