git-guardrails-claude-code

A setup guide for hooks that block an agent from running selected destructive Git commands, such as deleting branches, discarding changes, or forcefully resetting a repository.

In plain words
What is it for?
Use it to install project-wide or global Git protections for Claude Code. The blocked commands can be customized and the hook can be tested with a sample command.
Why use it?
It adds a safety check before risky Git operations run. This reduces the chance that an agent will permanently remove code or uncommitted work.

Command

Part of the dev-helpers plugin — 13 commands, 1 hook shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/refo/claude-dev-helpers/git-guardrails-claude-code
Clone the repo
git clone --depth 1 https://github.com/refo/claude-dev-helpers

Or install dev-helpers, the plugin that ships this one along with the rest of its 13 commands, 1 hook.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 256 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00256
Opus 5 $0.00000 $0.00128
Sonnet 5 $0.00000 $0.00051
Haiku 4.5 $0.00000 $0.00026

Measured 2d ago against content hash 373deb558a2a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

git-guardrails-claude-code scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

**Scope Selection**: First, determine whether to install guardrails for the current project only (`.claude/settings.json`) or globally across all projects (`~/.claude/settings.json`).
plugins/dev-helpers/commands/git-guardrails-claude-code.md · 18 lines

What it actually says

Git Guardrails for Claude Code

This guide enables safety hooks that prevent Claude from executing destructive git operations. The setup intercepts dangerous commands like git push, git reset --hard, git clean, git branch -D, and git checkout . before execution.

Installation Process

Scope Selection: First, determine whether to install guardrails for the current project only (.claude/settings.json) or globally across all projects (~/.claude/settings.json).

Hook Placement: Copy the blocking script from scripts/block-dangerous-git.sh to either .claude/hooks/block-dangerous-git.sh (project-level) or ~/.claude/hooks/block-dangerous-git.sh (global), then make it executable.

Configuration: Add a PreToolUse hook entry to your settings file that targets the Bash tool and references the script location using the appropriate path variable.

Customization: After installation, you can modify which git commands are blocked by editing the script directly.

Testing: Verify functionality by piping a sample git command through the script — it should return exit code 2 with a blocking message.

The hook prevents Claude from accessing these operations while remaining transparent about why commands cannot execute.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 18 lines · 0 tokens per session scan B 373deb558a2a

Subscribe to this mod's changes

git-guardrails-claude-code is a command published in the GitHub repository refo/claude-dev-helpers (2 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 256 tokens. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.