Claude Code Complete Usage Guide

Claude Code isn't another AI IDE. It's a CLI-native intelligent coding assistant. This guide is compiled from official docs and covers installation, configuration, MCP integration, security management, and everything in between.

---

Claude Code vs AI IDE Comparison

---

Installation

Option 1: Mirror Site (Recommended for China-based Users)

macOS:

curl -fsSL https://aicodewith.com/claudecode/resources/install-script-macos | bash

Linux:

curl -fsSL https://aicodewith.com/claudecode/resources/install-script-linux | bash

Windows (WSL):

# Install WSL2
wsl --install

# Install Ubuntu from Microsoft Store, then run
curl -fsSL https://aicodewith.com/claudecode/resources/install-script-wsl | bash

Option 2: NPM Official Install

# Global install
npm install -g @anthropic-ai/claude-code

# Verify version
claude --version

---

First-Time Setup

API Configuration (Mirror site users can skip)

# Get API key from the official site
export ANTHROPIC_API_KEY="sk-your-key-here"

# Pick your shell
# Bash
echo 'export ANTHROPIC_API_KEY="sk-your-key-here"' >> ~/.bashrc
source ~/.bashrc

# Zsh
echo 'export ANTHROPIC_API_KEY="sk-your-key-here"' >> ~/.zshrc
source ~/.zshrc

Basic Config

# Change default settings
claude config set -g model claude-sonnet-4
claude config set -g verbose true
claude config set -g outputFormat text

# Test installation
claude "Hello, Claude!"
claude /doctor

Security Settings (Optional)

export DISABLE_TELEMETRY=1                 # Don't send usage stats
export DISABLE_ERROR_REPORTING=1           # Don't send error logs
export DISABLE_NON_ESSENTIAL_MODEL_CALLS=1 # Save tokens

# Restrict tool permissions
claude config set allowedTools "Edit,View"

---

Basic Commands

Core Interaction

claude                           # Launch
claude "fix this bug"            # One-shot command
claude -p "<prompt>"             # Single print mode
cat file | claude -p "<prompt>"  # Large file reading
claude update                    # Update client
claude mcp                       # MCP wizard

Conversation Commands

claude -c              # Continue last conversation
claude -r <id>         # Resume by session ID
claude --resume <id>   # Resume long conversation
claude --resume <name> # Resume by name

Slash Commands

---

MCP Integration

MCP extends Claude's capabilities by connecting to external services, databases, and APIs.

Basic MCP Commands

claude mcp list              # List existing MCP services
claude mcp add <name> <cmd>  # Add new MCP
claude mcp remove <name>     # Remove MCP

Config file location: ~/.claude.json

Common MCP Services

Git MCP:

npm install -g git-mcp-server
claude mcp add git "git-mcp-server"
claude mcp add github "github-mcp-server --token $GITHUB_TOKEN"

Database MCP:

npm install -g postgres-mcp-server
export POSTGRES_URL="postgresql://user:password@localhost:5432/mydb"
claude mcp add postgres "postgres-mcp-server --url $POSTGRES_URL"

MCP Tool Permissions

# Grant specific MCP permissions
claude --allowedTools "mcp__git__commit,mcp__git__push"

# Wildcard permissions
claude --allowedTools "mcp__postgres__*"

# Combine MCP with built-in tools
claude --allowedTools "Edit,View,mcp__git__*"

---

Configuration System

Global Config

Location: ~/.claude.json

{
	"model": "claude-sonnet-4",
	"verbose": true,
	"outputFormat": "text",
	"allowedTools": ["Edit", "View"],
	"disallowedTools": []
}

Project Config

Location: .claude/settings.json

{
	"model": "claude-sonnet-4",
	"systemPrompt": "You are a senior developer working on this project",
	"allowedTools": ["Edit", "View", "Bash(git:*)", "Bash(npm:*)"]
}

Environment Variables

---

Security & Permission Management

Permission Levels

Tool Permission Management

# Grant specific tool permissions
claude --allowedTools "Edit,View"

# Scoped permissions (git only)
claude --allowedTools "Bash(git:*)"

# Multiple scopes
claude --allowedTools "Bash(git:*),Bash(npm:*)"

# Full permission mode (dangerous!)
claude --dangerously-skip-permissions

Recommended Security Strategies

1. Restrict to specific permissions

# Good: specific permissions
claude --allowedTools "Edit,View,Bash(git:status)"

# Bad: too broad
claude --allowedTools "Bash"

2. Protect sensitive data

# Good: use environment variables
export DATABASE_URL="postgresql://user:pass@host/db"

# Bad: hardcoded credentials
claude "connect to postgresql://user:password123@host/db"

3. Audit permissions regularly

claude config get allowedTools
claude config get disallowedTools
claude config list

---

Thinking Mode

Trigger deeper thinking with prompt keywords:

Example:

claude -p "We have a tricky concurrency bug. ultrathink and propose a fix."

• Keywords can appear anywhere in the prompt
• If multiple keywords exist, highest level wins

---

Config Command Reference

claude config list              # Show all current settings
claude config get <key>         # Show specific setting
claude config set -g <key> <value>  # Set global value
claude config add -g <key> <value>  # Append to array setting
claude config remove -g <key> <value>  # Remove from array

Note: -g flag means global setting. Without it, settings apply to current project only.

Common Config Keys

---

Team Collaboration & CI/CD

GitHub Actions Integration

name: Claude Code Review
on:
    pull_request:
        branches: [main, develop]

jobs:
    claude-review:
        runs-on: ubuntu-latest
        steps:
            - uses: actions/checkout@v4

            - name: Setup Node.js 18
              uses: actions/setup-node@v4
              with:
                  node-version: '18'

            - name: Install Claude Code
              run: npm install -g @anthropic-ai/claude-code

            - name: Review PR
              env:
                  ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
              run: |
                  claude -p "Review changes for security issues and bugs" \
                    --allowedTools "View" \
                    --output-format json > review-results.json

            - name: Upload results
              uses: actions/upload-artifact@v4
              with:
                  name: claude-review
                  path: review-results.json

Local Git Pre-commit Hook

#!/usr/bin/env bash
# .git/hooks/pre-commit (chmod +x)

staged=$(git diff --cached --name-only --diff-filter=ACM)
[ -z "$staged" ] && exit 0

payload=$(echo "$staged" | xargs cat)

analysis=$(echo "$payload" | \
  claude -p "Review these changes for issues before commit" \
    --allowedTools "View" \
    --output-format json)

if echo "$analysis" | jq -e '.critical_issues[]' >/dev/null 2>&1; then
  echo "❌ Critical issues found – commit blocked"
  exit 1
fi

echo "✅ Claude analysis passed"

---

Advanced Features

Persistent Memory (CLAUDE.md)

Generated by /init, lives at the project root:

# Project: My Application

## Overview

This is a React/Node.js application with PostgreSQL database.

## Architecture

-   Frontend: React 18 with TypeScript
-   Backend: Node.js with Express
-   Database: PostgreSQL 14

## Development Guidelines

-   Use TypeScript for all new code
-   Follow ESLint configuration
-   Write tests for new features

Memory Command

claude /memory           # Edit project memory
claude /memory view      # View current memory

Multi-Directory Workspace

# Add multiple directories
claude --add-dir ../frontend ../backend ../shared

# Project-level analysis
claude "analyze the entire application architecture"

---

Summary

Claude Code is a powerful CLI coding assistant. Core strengths:

• Native Claude 4 - completely uncut
• Unlimited tool calls - complex tasks run to completion
• 200K+ context - sees the entire project at once
• Fully autonomous agent - start-to-finish execution
• MCP extensions - connect external services and databases
• CI/CD integration - automated code review

Use this guide as a reference -- look up the relevant section when you need it.

---

Original source: Zhihu @Cluade2Claude | Compiled 2025-10-18