safety-help

Audit Claude Code sandbox, hooks, deny rules, and Gitleaks configurations.

12|2|Updated Feb 27, 2026
One-click install
npx skills add https://github.com/hartphoenix/weft --skill safety-help
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: safety-help
Source: https://github.com/hartphoenix/weft/tree/main/.claude/skills/safety-help
Command: npx skills add https://github.com/hartphoenix/weft --skill safety-help

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps users audit and verify their security setup for Claude Code to prevent vulnerabilities and misconfigurations.

Core Features & Use Cases

  • Security Configuration Audit: Checks sandbox settings, hooks, deny rules, and Gitleaks integrations for compliance with best practices.
  • Detailed Reporting: Surfaces findings in plain language with tailored recommendations to fix issues.
  • Use Case: A user wants to ensure their environment is protected before deploying sensitive code, and runs /safety-help to identify potential security gaps.

Quick Start

Invoke /safety-help in your Claude environment to initiate an audit of your security setup and receive a detailed report on vulnerabilities and recommended actions.

Frequently Asked Questions about safety-help

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit my Claude Code security configuration?

To audit your Claude Code security configuration, run the `/safety-help` command in your environment. The Skill analyzes your sandbox, hooks, deny rules, and Gitleaks integrations to surface misconfigurations and provide tailored recommendations.

What is checked during a Claude Code security audit?

A Claude Code security audit checks sandbox settings, hooks, deny rules, and Gitleaks pre-commit/pre-push hooks. It identifies weaknesses, risky settings, and deviations from secure coding best practices.

How do I integrate Gitleaks pre-commit hooks with Claude Code?

To verify Gitleaks pre-commit and pre-push hooks with Claude Code, use `/safety-help` to analyze your local hook files. The audit checks your integration for compliance and reports any risky settings or misconfigurations.

Do I need local settings.json access to audit Claude Code security?

Yes, auditing Claude Code security requires access to your local settings.json and hook files. The Skill analyzes these specific files to accurately assess your sandbox configuration and deny rules.

Why are my Claude Code hooks creating security vulnerabilities?

Your Claude Code hooks may have security vulnerabilities due to misconfigurations or risky settings in your local files. Running `/safety-help` analyzes your hook files and provides plain-language recommendations to fix these gaps.