configuration

Configure Claude Code hierarchical settings and permission rules.

5|1|Updated Oct 30, 2025
One-click install
npx skills add https://github.com/reggiechan74/claude-plugins --skill configuration
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: configuration
Source: https://github.com/reggiechan74/claude-plugins/tree/main/plugins/claude-code-metaskill/skills/configuration
Command: npx skills add https://github.com/reggiechan74/claude-plugins --skill configuration

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Managing AI behavior, tool access, and environment variables can be complex and inconsistent across projects or teams. This Skill centralizes and clarifies Claude Code's hierarchical configuration, ensuring consistent and secure operations.

Core Features & Use Cases

  • Hierarchical Settings: Configure settings globally, per-project, or via enterprise policies using settings.json files.
  • Granular Permissions: Control tool and file access with allow, deny, and ask patterns for enhanced security.
  • Environment Variables: Easily set critical variables like ANTHROPIC_API_KEY or MAX_THINKING_TOKENS for consistent session behavior.
  • Use Case: Define a project-level settings.json to ensure all team members use specific models, restrict access to sensitive files, and auto-approve common build commands, standardizing your team's workflow.

Quick Start

Show me how to set up project-level permissions to deny access to .env files.

Frequently Asked Questions about configuration

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

FAQPage Schema
How do I set up hierarchical configuration for Claude Code across my team?

Hierarchical configuration uses settings.json files at user, project, and enterprise levels with precedence rules. Project-level settings override user defaults, and enterprise policies apply organization-wide, ensuring consistent behavior across your team's workflows.

How do I control file and tool access with permissions in Claude Code?

Permissions use allow, deny, and ask patterns to grant or restrict access to files and tools. Configure these in settings.json to auto-approve common commands, block sensitive files like .env, or prompt for approval on risky operations.

What's the best way to manage environment variables like ANTHROPIC_API_KEY in Claude Code?

Set environment variables in settings.json or via slash commands to persist critical values across sessions. This centralizes configuration, prevents hardcoding secrets, and ensures consistent variable behavior in your Claude Code environment.

Can I enforce project-level settings that apply to all team members?

Yes, project-level settings.json files enforce consistent configuration across your team. Place settings in your project root to standardize model choices, restrict file access, and auto-approve build commands, ensuring uniform security and behavior.

How do sandbox controls work in Claude Code configuration?

Sandbox controls restrict Claude Code's system access and file operations based on your permission rules. Configure sandbox settings in settings.json to isolate sensitive directories, limit tool execution, and control what Claude Code can read or modify.

Can I use slash commands to interactively configure Claude Code settings?

Yes, interactive slash commands let you configure Claude Code settings on demand without editing files. Use them to toggle permissions, set environment variables, or adjust sandbox behavior for immediate changes during your session.