Threads - https://www.threads.com/@george_sl_liu
BlueSky - https://bsky.app/profile/georgesl.bsky.social

My Claude Code Setup

A comprehensive starter kit for Claude Code projects featuring a memory bank system for context retention, pre-configured hooks, skills, subagents, and MCP server recommendations.

You can also use Claude Code with Z.AI to get higher token usage quotas and get access to Z.AI GLM-4.7 LLM models within Claude Code. Use Z.AI invite code for additional 10% discount which can stack with current 50-60% yearly discounts.

Quick Start
Beginner
- 1. Getting Started
- 2. Memory Bank System
Intermediate
- 3. Extending Claude Code
- 4. Development Workflows
Advanced
Reference
- 8. Configuration Reference
- 9. Resources

Quick Start

Get up and running in under 5 minutes:

Prerequisites Checklist:

Paid Claude AI account (Pro $20/mo, Max $100/mo, or Max $200/mo)
Node.js 18+ installed
Git installed
macOS: brew install ripgrep fd jq

Installation Options:

Choose the method that works best for you:

Option A: Clone Entire Repository (New Projects)

# 1. Clone this repository as your new project
git clone https://github.com/centminmod/my-claude-code-setup.git my-project
cd my-project

# 2. Remove template README files (create your own project README)
rm README.md README-v2.md README-v3.md README-v4.md

# 3. Reinitialize git for your own project (optional)
rm -rf .git
git init

# 4. Launch Claude Code and initialize
claude
# Then run: /init

Option B: Selective Copy (Existing Projects)

Copy only the files you need into your existing project:

# Core files (recommended)
cp /path/to/my-claude-code-setup/CLAUDE.md your-project/
cp -r /path/to/my-claude-code-setup/.claude your-project/

# Or selectively copy specific components:
cp -r /path/to/my-claude-code-setup/.claude/commands your-project/.claude/
cp -r /path/to/my-claude-code-setup/.claude/skills your-project/.claude/
cp -r /path/to/my-claude-code-setup/.claude/agents your-project/.claude/
cp -r /path/to/my-claude-code-setup/.claude/hooks your-project/.claude/

Option C: Download Specific Files (GitHub)

Browse the repository on GitHub and download individual files:

CLAUDE.md - Memory bank main file
.claude/settings.json - Settings template
.claude/commands/ - Slash commands you want
.claude/skills/ - Skills you want
.claude/agents/ - Subagents you want

After copying files, launch Claude Code in your project and run /init.

BEGINNER SECTION

1. Getting Started

1.1 What This Repository Provides

Memory Bank System: Structured context files for persistent memory across sessions
Pre-configured Settings: Optimized .claude/settings.json with fast tools
Custom Hooks: Desktop notifications when Claude Code completes tasks
Custom Skills: Specialized agents for documentation lookup
Custom Subagents: Task-specific agents for code search, UX design, and more
Custom Slash Commands: Security audits, refactoring plans, prompt engineering
MCP Server Recommendations: Curated list of useful MCP servers

1.2 Prerequisites

Requirement	Version	Purpose
Claude AI Account	Pro/Max	Required for Claude Code access
Node.js	18+	Runtime environment
Git	Latest	Version control
ripgrep	Latest	Fast content search
fd	Latest	Fast file finding
jq	Latest	JSON processing

Operating System Support:

macOS: Fully supported
Linux: Fully supported
Windows: Supported (PowerShell or CMD)

1.3 Installation Steps

Copy Repository Files

Copy all files from this GitHub repo to your project directory (where your intended codebase will be).
Modify Template Files

Edit CLAUDE.md to match your project. The .claude/settings.json file includes Terminal-Notifier for macOS notifications (setup guide). Remove this file if not using macOS.
Initialize Memory Bank

After launching Claude Code in your project directory, run /init to analyze your codebase and populate memory bank files.
Install Fast Tools (macOS)
```
brew install ripgrep fd jq
```

1.4 Optional Enhancements

Visual Studio Code: Install VS Code and the Claude Code Extension
- Beginner YouTube guide
- Additional guide
GitHub Integration: Set up Git for VS Code
- Git setup guide
- VS Code + Git tutorial
Dev Container: Run Claude Code, Codex GPT-5, Gemini CLI, and more in an isolated VS Code dev container with full YOLO mode support.
Platform-Specific Docs: If using Cloudflare/ClerkOS, keep CLAUDE-cloudflare.md or CLAUDE-cloudflare-mini.md and reference them in CLAUDE.md. For Convex database, use CLAUDE-convex.md.

1.5 Useful Learning Resources

Advent of Claude: 31 Days of Claude Code by Ado Kukic (Anthropic)
Claude Code with Opus 4.5 by Alex Finn
Claude Code Overview by Matt Maher

2. Memory Bank System

2.1 What is Memory Bank?

The Memory Bank System is a structured set of markdown files that help Claude Code retain context across multiple chat sessions. Instead of starting fresh each time, Claude Code reads these files to understand your project's patterns, decisions, and current state.

2.2 Core Context Files

File	Purpose
`CLAUDE.md`	Main entry point with project overview and AI guidance
`CLAUDE-activeContext.md`	Current session state, goals, and progress
`CLAUDE-patterns.md`	Established code patterns and conventions
`CLAUDE-decisions.md`	Architecture decisions and rationale (ADRs)
`CLAUDE-troubleshooting.md`	Common issues and proven solutions
`CLAUDE-config-variables.md`	Configuration variables reference
`CLAUDE-temp.md`	Temporary scratch pad (only read when referenced)

2.3 Setting Up Memory Bank

Copy the template files from this repository to your project root
Run /init in Claude Code to analyze your codebase
Claude Code will populate the memory bank files based on your project

2.4 Updating Memory Bank

After completing significant work, ask Claude Code to update the memory bank:

update memory bank

2.5 Memory Bank Example Output

Here's what Claude Code's thinking looks like when updating the memory bank:

> update memory bank

∴ Thinking…

  The user wants me to update the memory bank files. I should update the relevant CLAUDE-*.md files to reflect the v0.1.6 changes and current state. Let me check
   which memory bank files exist and update them.

∴ Thinking…

  I need to continue updating the memory bank files. I'll update:
  1. CLAUDE-patterns.md - Add export path handling pattern (Pattern 22)
  2. CLAUDE-decisions.md - Add ADR-017 for Phoneme API migration and ADR-018 for export path handling
  3. CLAUDE-troubleshooting.md - Add new issues and solutions for v0.1.6

∴ Thinking…

  The memory bank files have been modified but not committed. The v0.1.6 release was already committed and tagged, but the memory bank documentation updates are
  still uncommitted. Let me provide a summary of the completed work and offer to commit these documentation changes.

⏺ Summary: v0.1.6 Release Complete ✅

  Memory Bank Updates Complete:
  - CLAUDE-activeContext.md - v0.1.6 session context
  - CLAUDE-patterns.md - Added patterns 22-24 (Export Path, Format Detection, API Migration)
  - CLAUDE-decisions.md - Added ADR-017 & ADR-018
  - CLAUDE-troubleshooting.md - Added 4 new issue/fix entries

  v0.1.6 Fixes Applied:
  1. Export to absolute paths (e.g., --export /tmp/test.wav)
  2. Format detection from filename extension (prevents .mp3.wav)
  3. Phoneme API migrated to /dev/phonemize
  4. Test scripts updated for --export "" syntax

  Test Results: All 29 tests passing

  The memory bank documentation updates are ready to commit. Would you like me to commit them

INTERMEDIATE SECTION

3. Extending Claude Code

3.1 Plugin System

Claude Code supports plugins from both official and third-party marketplaces.

Official Anthropic Marketplace

Browse and install plugins from the official marketplace.

# Install a plugin
/plugin install plugin-name@claude-plugins-official

# Update marketplace
/plugin marketplace update claude-plugins-official

Recommended Official Plugins

Plugin	Description	Install Command
code-simplifier	Simplifies and refines code for clarity, consistency, and maintainability	`/plugin install code-simplifier`
frontend-design	Creates distinctive, production-grade frontend interfaces	`/plugin install frontend-design@claude-code-plugins`
feature-dev	Systematic 7-phase approach to building new features	`/plugin install feature-dev@claude-code-plugins`
ralph-wiggum	Interactive self-referential AI loops for iterative development	`/plugin install ralph-wiggum@claude-code-plugins`

Note: Ralph Wiggum may have issues on some systems. See #16398 and #16389. Usage guide: YouTube by Greg Isenberg | GitHub

Third-Party Marketplaces

Plugin	Source	Description	Install Commands
safety-net	cc-marketplace	Catches destructive git/filesystem commands before execution	`/plugin marketplace add kenryu42/cc-marketplace` `/plugin install safety-net@cc-marketplace`
glm-plan-usage	zai-coding-plugins	Query Z.AI usage statistics	`/plugin marketplace add zai/zai-coding-plugins` `/plugin install glm-plan-usage@zai-coding-plugins`
Cloudflare Skills	cloudflare/skills	Development skills for Cloudflare platform (Workers, Agents SDK)	`/plugin marketplace add cloudflare/skills`

3.2 MCP Servers

MCP (Model Context Protocol) servers extend Claude Code's capabilities by connecting to external tools and documentation.

Recommended MCP Servers

Server	Purpose	Transport	Token Cost	Install Command
Context7	Documentation lookup for any library	SSE	Low	See below
Cloudflare Docs	Cloudflare documentation	SSE	Low	See below
Usage Metrics	Claude Code usage and cost tracking	stdio	Low	See below
Gemini CLI	Access to Gemini models	stdio	Variable	See below
Notion	Notion workspace integration	stdio	Variable	See below
Chrome DevTools	Browser automation & debugging	stdio	~17K tokens	See below

Installation Commands

Context7 MCP (with API key):

claude mcp add --transport http context7 https://mcp.context7.com/mcp --header "CONTEXT7_API_KEY: YOUR_API_KEY" -s user

Cloudflare Documentation MCP:

claude mcp add --transport sse cf-docs https://docs.mcp.cloudflare.com/sse -s user

Usage Metrics MCP:

claude mcp add --transport stdio metrics -s user -- uv run --directory /path/to/your/mcp-server metrics-server

Gemini CLI MCP:

claude mcp add gemini-cli /path/to/.venv/bin/python /path/to/mcp_server.py -s user -e GEMINI_API_KEY='YOUR_KEY' -e OPENROUTER_API_KEY='YOUR_KEY'

Notion MCP:

claude mcp add-json notionApi '{"type":"stdio","command":"npx","args":["-y","@notionhq/notion-mcp-server"],"env":{"OPENAPI_MCP_HEADERS":"{\"Authorization\": \"Bearer ntn_API_KEY\", \"Notion-Version\": \"2022-06-28\"}"}}' -s user

Chrome DevTools MCP (high token cost - use on-demand):

Due to the ~17K token overhead across 26 tools, install only when needed via --mcp-config:

claude --mcp-config .claude/mcp/chrome-devtools.json

Create .claude/mcp/chrome-devtools.json:

{
  "mcpServers": {
    "chrome-devtools": {
      "command": "npx",
      "args": ["-y", "chrome-devtools-mcp@latest"]
    }
  }
}

Verify MCP Servers

claude mcp list
# Output:
# context7: https://mcp.context7.com/sse (SSE) - ✓ Connected
# cf-docs: https://docs.mcp.cloudflare.com/sse (SSE) - ✓ Connected
# metrics: uv run --directory /path/to/mcp-server metrics-server - ✓ Connected

Usage Metrics Example

The get_current_cost MCP tool returns today's total USD cost:

{
  "metric": "Total Cost Today",
  "value": 27.149809833783127,
  "formatted": "$27.1498",
  "unit": "currencyUSD"
}

4. Development Workflows

4.1 Git Worktrees for Parallel Sessions

Git worktrees allow you to run parallel Claude Code sessions with complete code isolation. Each worktree has its own working directory while sharing Git history.

Benefits:

Run multiple AI coding sessions in parallel
Each worktree has independent file state
Changes in one worktree won't affect others
Ideal for experimental features or YOLO mode

Official Documentation: Run parallel Claude Code sessions with git worktrees

macOS / Linux Shell Functions

Add to ~/.bashrc, ~/.zshrc, or ~/.bash_aliases:

# Codex CLI worktree launcher
cx() {
    local branch_name
    if [ -z "$1" ]; then
        branch_name="worktree-$(date +%Y%m%d-%H%M%S)"
    else
        branch_name="$1"
    fi
    git worktree add "../$branch_name" -b "$branch_name" && \
    cd "../$branch_name" || return 1
    codex -m gpt-5-codex --config model_reasoning_effort='xhigh'
}

# Claude Code worktree launcher
clx() {
    local branch_name
    if [ -z "$1" ]; then
        branch_name="worktree-$(date +%Y%m%d-%H%M%S)"
    else
        branch_name="$1"
    fi
    git worktree add "../$branch_name" -b "$branch_name" && \
    cd "../$branch_name" || return 1
    claude --model opusplan --permission-mode plan
}

After adding, reload: source ~/.bashrc or source ~/.zshrc

Windows PowerShell Functions

Add to your PowerShell profile (notepad $PROFILE):

# Codex CLI worktree launcher
function cx {
    param([string]$BranchName)
    if (-not $BranchName) {
        $BranchName = "worktree-$(Get-Date -Format 'yyyyMMdd-HHmmss')"
    }
    git worktree add "../$BranchName" -b $BranchName
    if ($LASTEXITCODE -eq 0) {
        Set-Location "../$BranchName"
        codex -m gpt-5-codex --config model_reasoning_effort='xhigh'
    }
}

# Claude Code worktree launcher
function clx {
    param([string]$BranchName)
    if (-not $BranchName) {
        $BranchName = "worktree-$(Get-Date -Format 'yyyyMMdd-HHmmss')"
    }
    git worktree add "../$BranchName" -b $BranchName
    if ($LASTEXITCODE -eq 0) {
        Set-Location "../$BranchName"
        claude --model opusplan --permission-mode plan
    }
}

Reload: . $PROFILE

Windows CMD Batch Files

Create in a directory in your PATH (e.g., C:\Users\YourName\bin\):

cx.bat - Codex CLI launcher:

@echo off
setlocal enabledelayedexpansion
if "%~1"=="" (
    for /f "tokens=2 delims==" %%I in ('wmic os get localdatetime /value') do set datetime=%%I
    set branch_name=worktree-!datetime:~0,8!-!datetime:~8,6!
) else (
    set branch_name=%~1
)
git worktree add "../%branch_name%" -b "%branch_name%"
if %errorlevel% equ 0 (
    cd "../%branch_name%"
    codex -m gpt-5-codex --config model_reasoning_effort='xhigh'
)
endlocal

clx.bat - Claude Code launcher:

@echo off
setlocal enabledelayedexpansion
if "%~1"=="" (
    for /f "tokens=2 delims==" %%I in ('wmic os get localdatetime /value') do set datetime=%%I
    set branch_name=worktree-!datetime:~0,8!-!datetime:~8,6!
) else (
    set branch_name=%~1
)
git worktree add "../%branch_name%" -b "%branch_name%"
if %errorlevel% equ 0 (
    cd "../%branch_name%"
    claude --model opusplan --permission-mode plan
)
endlocal

Usage

# Create worktree with custom name
clx feature-auth
cx bugfix-123

# Create worktree with auto-generated timestamp name
clx
cx

Worktree Management

# List all worktrees
git worktree list

# Remove a worktree when done
git worktree remove ../worktree-name

# Clean up stale worktree references
git worktree prune

Environment Setup

Each new worktree needs its own development environment:

JavaScript/Node.js: Run npm install or yarn
Python: Create virtual environment or run pip install -r requirements.txt
Other languages: Follow your project's standard setup process

The .worktreeinclude File

When Claude Code creates a worktree, files ignored via .gitignore aren't automatically available. The .worktreeinclude file specifies which ignored files should be copied.

How It Works:

Uses .gitignore-style patterns
Only files matched by both .worktreeinclude AND .gitignore are copied

Create .worktreeinclude in your repository root:

# Environment files
.env
.env.local
.env.*

# Claude Code local settings
**/.claude/settings.local.json

Common Use Cases:

.env files with API keys and secrets
.env.local for local development overrides
.claude/settings.local.json for personal Claude Code settings

Claude Desktop Worktree Location

When using Claude Code via the Claude Desktop app:

Default location: ~/.claude-worktrees
Configurable through Claude Desktop app settings
Repository must be Git initialized for worktree sessions

Official Documentation: Claude Code on Desktop

Local Ignores Without .gitignore

Use .git/info/exclude to ignore files locally without modifying shared .gitignore:

# Edit the local exclude file
nano .git/info/exclude

Add patterns:

# Local IDE settings
.idea/
*.swp

# Personal scripts
my-local-scripts/

File	Scope	Committed to Git
`.gitignore`	Shared with team	Yes
`.git/info/exclude`	Local only	No
`~/.config/git/ignore`	Global (all repos)	No

4.2 Status Lines

Configure custom status lines in ~/.claude/settings.json:

{
  "statusLine": {
    "type": "command",
    "command": "~/.claude/statuslines/statusline.sh",
    "padding": 0
  }
}

Example Status Line Script

Create ~/.claude/statuslines/statusline.sh:

#!/bin/bash
# Read JSON input from stdin
input=$(cat)

# Extract model and workspace values
MODEL_DISPLAY=$(echo "$input" | jq -r '.model.display_name')
CURRENT_DIR=$(echo "$input" | jq -r '.workspace.current_dir')

# Extract context window metrics
INPUT_TOKENS=$(echo "$input" | jq -r '.context_window.total_input_tokens')
OUTPUT_TOKENS=$(echo "$input" | jq -r '.context_window.total_output_tokens')
CONTEXT_SIZE=$(echo "$input" | jq -r '.context_window.context_window_size')

# Extract cost metrics
COST_USD=$(echo "$input" | jq -r '.cost.total_cost_usd')
LINES_ADDED=$(echo "$input" | jq -r '.cost.total_lines_added')
LINES_REMOVED=$(echo "$input" | jq -r '.cost.total_lines_removed')

# Extract percentage metrics
USED_PERCENTAGE=$(echo "$input" | jq -r '.context_window.used_percentage')
REMAINING_PERCENTAGE=$(echo "$input" | jq -r '.context_window.remaining_percentage')

# Format tokens as Xk
format_tokens() {
    local num="$1"
    if [ "$num" -ge 1000 ]; then
        echo "$((num / 1000))k"
    else
        echo "$num"
    fi
}

# Generate progress bar for context usage
generate_progress_bar() {
    local percentage=$1
    local bar_width=20
    local filled=$(awk "BEGIN {printf \"%.0f\", ($percentage / 100) * $bar_width}")
    local empty=$((bar_width - filled))
    local bar=""
    for ((i=0; i<filled; i++)); do bar+="█"; done
    for ((i=0; i<empty; i++)); do bar+="░"; done
    echo "$bar"
}

# Calculate total
TOTAL_TOKENS=$((INPUT_TOKENS + OUTPUT_TOKENS))

# Generate progress bar
PROGRESS_BAR=$(generate_progress_bar "$USED_PERCENTAGE")

# Show git branch if in a git repo
GIT_BRANCH=""
if git -C "$CURRENT_DIR" rev-parse --git-dir > /dev/null 2>&1; then
    BRANCH=$(git -C "$CURRENT_DIR" branch --show-current 2>/dev/null)
    if [ -n "$BRANCH" ]; then
        # Worktree detection
        GIT_DIR=$(git -C "$CURRENT_DIR" rev-parse --git-dir 2>/dev/null)
        WORKTREE=""
        if [[ "$GIT_DIR" == *".git/worktrees/"* ]] || [[ -f "$GIT_DIR/gitdir" ]]; then
            WORKTREE=" 🌳"
        fi
        # Ahead/behind detection
        AHEAD_BEHIND=""
        UPSTREAM=$(git -C "$CURRENT_DIR" rev-parse --abbrev-ref '@{u}' 2>/dev/null)
        if [ -n "$UPSTREAM" ]; then
            AHEAD=$(git -C "$CURRENT_DIR" rev-list --count '@{u}..HEAD' 2>/dev/null || echo 0)
            BEHIND=$(git -C "$CURRENT_DIR" rev-list --count 'HEAD..@{u}' 2>/dev/null || echo 0)
            if [ "$AHEAD" -gt 0 ] && [ "$BEHIND" -gt 0 ]; then
                AHEAD_BEHIND=" ↕${AHEAD}/${BEHIND}"
            elif [ "$AHEAD" -gt 0 ]; then
                AHEAD_BEHIND=" ↑${AHEAD}"
            elif [ "$BEHIND" -gt 0 ]; then
                AHEAD_BEHIND=" ↓${BEHIND}"
            fi
        fi
        GIT_BRANCH=" | 🌿 $BRANCH${WORKTREE}${AHEAD_BEHIND}"
    fi
fi

echo "[$MODEL_DISPLAY] 📁 ${CURRENT_DIR##*/}${GIT_BRANCH}
Context: [$PROGRESS_BAR] ${USED_PERCENTAGE}%
Cost: \$${COST_USD} | +${LINES_ADDED} -${LINES_REMOVED} lines"

4.3 IDE Integration

VS Code Extension: Claude Code for VS Code
Dev Containers: Isolated dev environment with Claude Code, Codex CLI, Gemini CLI, and YOLO mode support

ADVANCED SECTION

5. Customization

5.1 Custom Subagents

Subagents are specialized tools that handle complex, multi-step tasks autonomously with their own context window and custom prompts. Learn more in the official documentation.

Included Subagents

Agent	Location	Purpose
memory-bank-synchronizer	`.claude/agents/`	Synchronizes memory bank documentation with codebase state
code-searcher	`.claude/agents/`	Efficient codebase navigation with optional Chain of Draft mode
get-current-datetime	`.claude/agents/`	Accurate Brisbane timezone (GMT+10) values
ux-design-expert	`.claude/agents/`	Comprehensive UX/UI design guidance with Tailwind CSS & Highcharts
zai-cli	`.claude/agents/`	CLI wrapper for z.ai GLM 4.7 model (used by consult-zai skill)
codex-cli	`.claude/agents/`	CLI wrapper for OpenAI Codex GPT-5.2 (used by consult-codex skill)

memory-bank-synchronizer

Proactively maintains consistency between CLAUDE-*.md files and source code:

Pattern documentation synchronization
Architecture decision updates
Technical specification alignment
Implementation status tracking

code-searcher

Efficiently searches the codebase with two modes:

Standard mode: Full detailed analysis
CoD mode: ~80% fewer tokens with ultra-concise responses

# Standard: "Find the payment processing code"
# CoD: "Find the payment processing code using CoD"

get-current-datetime

Returns accurate Brisbane, Australia (GMT+10) timestamps:

Supports multiple formats (default, filename, readable, ISO)
Eliminates timezone confusion

ux-design-expert

Comprehensive UX/UI guidance:

UX flow optimization
Premium UI design with Tailwind CSS
Data visualization with Highcharts
Accessibility compliance

zai-cli

CLI wrapper for z.ai GLM 4.7 model:

Executes z.ai CLI with JSON output
Used internally by consult-zai skill
Returns raw output for skill to process

codex-cli

CLI wrapper for OpenAI Codex GPT-5.2:

Executes Codex CLI in readonly mode
Used internally by consult-codex skill
Returns raw output for skill to process

5.2 Custom Skills

Skills provide specialized capabilities. See Agent Skills documentation.

Included Skills

Skill	Purpose	Invocation
claude-docs-consultant	Selectively fetches official Claude Code documentation from docs.claude.com	Automatic when working on Claude Code features
consult-zai	Dual-AI consultation comparing z.ai GLM 4.7 and code-searcher	`/consult-zai "question"` or via Skill tool
consult-codex	Dual-AI consultation comparing Codex GPT-5.2 and code-searcher	`/consult-codex "question"` or via Skill tool
ai-image-creator	Generate PNG images using AI (multiple models including Gemini, FLUX.2, Riverflow, SeedDream, GPT-5, GPT-5.4 Image 2 via OpenRouter/Cloudflare AI Gateway). Supports transparent backgrounds (`-t`), reference image editing (`-r`), image analysis and description (`--analyze`), per-project cost tracking (`--costs`), prompt enhancement patterns, and composite banner generation for multi-size logo banners via ImageMagick. Requires setup	`/ai-image-creator` or via Skill tool
session-metrics	Parse Claude Code JSONL logs into per-turn token/cost/cache breakdowns with multi-format export (text/JSON/CSV/MD/HTML). Includes 5-hour session blocks with 7/14/30-day counters for weekly-cap tracking, weekly roll-up with percentage deltas, session duration + burn rate, hour-of-day bars, weekday × hour punchcard, and pluggable chart libraries via `--chart-lib {highcharts,uplot,chartjs,none}` (Highcharts default, uPlot/Chart.js MIT alternatives, all vendored + SHA-256 verified). Instance-wide dashboard (`--all-projects`, v1.14.0+) rolls up every project under `~/.claude/projects/` into one dated export folder with per-project HTML drilldowns and a daily-cost timeline. Zero network at runtime, stdlib-only Python.	Auto-triggers on cost/usage questions or `uv run python .claude/skills/session-metrics/scripts/session-metrics.py` — also installable via `/plugin marketplace add centminmod/claude-plugins` + `/plugin install session-metrics@centminmod`

5.3 Custom Slash Commands

`/anthropic` Commands

Command	Purpose
`/apply-thinking-to`	Applies Anthropic's extended thinking patterns to enhance prompts
`/convert-to-todowrite-tasklist-prompt`	Converts prompts to TodoWrite tasklist methods (60-70% speed improvement)
`/update-memory-bank`	Updates CLAUDE.md and memory bank files

`/ccusage` Commands

Command	Purpose
`/ccusage-daily`	Generates comprehensive usage cost analysis with daily breakdowns

`/cleanup` Commands

Command	Purpose
`/cleanup-context`	Memory bank optimization (15-25% token reduction)

`/documentation` Commands

Command	Purpose
`/create-readme-section`	Generates README sections with professional formatting
`/create-release-note`	Creates dual release notes (customer-facing + technical)

`/security` Commands

Command	Purpose
`/security-audit`	Comprehensive security audit using OWASP guidelines
`/check-best-practices`	Analyzes code against language-specific best practices
`/secure-prompts`	Detects prompt injection attacks and malicious instructions

Example test prompts at .claude/commands/security/test-examples/

`/architecture` Commands

Command	Purpose
`/explain-architecture-pattern`	Identifies and explains architectural patterns

`/promptengineering` Commands

Command	Purpose
`/convert-to-test-driven-prompt`	Transforms requests into TDD-style prompts
`/batch-operations-prompt`	Optimizes prompts for parallel processing

`/refactor` Commands

Command	Purpose
`/refactor-code`	Creates detailed refactoring plans with risk assessment

5.4 Hooks System

Hooks run custom commands before or after tool execution.

Included Hooks

The STOP hook uses Terminal-Notifier for macOS desktop notifications when Claude Code completes a response. Setup: Terminal-Notifier

6. Alternative Providers

6.1 Using Z.AI with Claude Code

Z.AI's GLM Coding Plan provides cost-effective access to high-performance language models optimized for coding.

10% Discount: Use invite code WWB8IFLROM

Overview

Performance: 55+ tokens/second for real-time interaction
Features: Vision Understanding, Web Search, Web Reader MCP servers
Multimodal: GLM-4.7 with state-of-the-art reasoning capabilities

Pricing & Plans

Plan	Prompts/5hrs	Approx. Monthly Cost	vs Claude
Lite	~120	~$3	3× Claude Pro quota
Pro	~600	Higher	3× Claude Max 5x quota
Max	~2,400	Higher	3× Claude Max 20x quota

Each prompt allows 15–20 model calls, yielding billions of tokens monthly at ~1% of standard API pricing.

Privacy & Data Handling

Data Location: Singapore
Privacy Guarantee: Z.AI does not store any content you provide or generate
See Privacy Policy

Prerequisites

Node.js 18+
Z.AI API key from Z.AI dashboard
Official documentation

Setup Instructions

Option 1: Automated Setup (macOS/Linux only)

curl -O "https://cdn.bigmodel.cn/install/claude_code_zai_env.sh" && bash ./claude_code_zai_env.sh

Option 2: Manual Configuration

Edit ~/.claude/settings.json:

{
  "env": {
    "ANTHROPIC_AUTH_TOKEN": "your-zai-api-key",
    "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
    "API_TIMEOUT_MS": "3000000"
  }
}

Shell Function Launchers

The easiest way to use Z.AI without affecting your existing Claude Code setup.

macOS / Linux (Bash/Zsh)

Add to ~/.bashrc, ~/.zshrc, or ~/.bash_aliases:

# Z.AI + Claude Code launcher
zai() {
    export ANTHROPIC_AUTH_TOKEN="your-zai-api-key"
    export ANTHROPIC_BASE_URL="https://api.z.ai/api/anthropic"
    export API_TIMEOUT_MS="3000000"
    claude "$@"
}

Reload: source ~/.bashrc or source ~/.zshrc

Windows PowerShell

Add to PowerShell profile (notepad $PROFILE):

# Z.AI + Claude Code launcher
function zai {
    $env:ANTHROPIC_AUTH_TOKEN = "your-zai-api-key"
    $env:ANTHROPIC_BASE_URL = "https://api.z.ai/api/anthropic"
    $env:API_TIMEOUT_MS = "3000000"
    claude $args
}

Reload: . $PROFILE

Windows CMD Batch

Create zai.bat in a directory in your PATH:

@echo off
set ANTHROPIC_AUTH_TOKEN=your-zai-api-key
set ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
set API_TIMEOUT_MS=3000000
claude %*

Model Mapping

GLM Coding Plan maps Claude model names to GLM models:

Claude Model	GLM Model
Opus	GLM-4.7
Sonnet	GLM-4.7
Haiku	GLM-4.5-Air

Customizing Model Mappings (optional):

In ~/.claude/settings.json:

{
  "env": {
    "ANTHROPIC_AUTH_TOKEN": "your-zai-api-key",
    "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
    "API_TIMEOUT_MS": "3000000",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "GLM-4.7",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "GLM-4.5",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "GLM-4.5-Air"
  }
}

Or in shell function:

zai() {
    export ANTHROPIC_AUTH_TOKEN="your-zai-api-key"
    export ANTHROPIC_BASE_URL="https://api.z.ai/api/anthropic"
    export API_TIMEOUT_MS="3000000"
    export ANTHROPIC_DEFAULT_OPUS_MODEL="GLM-4.7"
    export ANTHROPIC_DEFAULT_SONNET_MODEL="GLM-4.5"
    export ANTHROPIC_DEFAULT_HAIKU_MODEL="GLM-4.5-Air"
    claude "$@"
}

Windows PowerShell Model Mapping

function zai {
    $env:ANTHROPIC_AUTH_TOKEN = "your-zai-api-key"
    $env:ANTHROPIC_BASE_URL = "https://api.z.ai/api/anthropic"
    $env:API_TIMEOUT_MS = "3000000"
    $env:ANTHROPIC_DEFAULT_OPUS_MODEL = "GLM-4.7"
    $env:ANTHROPIC_DEFAULT_SONNET_MODEL = "GLM-4.5"
    $env:ANTHROPIC_DEFAULT_HAIKU_MODEL = "GLM-4.5-Air"
    claude $args
}

Windows CMD Model Mapping

@echo off
set ANTHROPIC_AUTH_TOKEN=your-zai-api-key
set ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
set API_TIMEOUT_MS=3000000
set ANTHROPIC_DEFAULT_OPUS_MODEL=GLM-4.7
set ANTHROPIC_DEFAULT_SONNET_MODEL=GLM-4.5
set ANTHROPIC_DEFAULT_HAIKU_MODEL=GLM-4.5-Air
claude %*

Usage

# Launch with Z.AI configuration
zai

# Launch with specific model
zai --model sonnet

# Launch with permission mode
zai --model opus --permission-mode plan

Z.AI + Git Worktree Integration

Combine Z.AI with git worktrees for isolated parallel sessions.

macOS / Linux (Bash/Zsh)

# Z.AI + Claude Code worktree launcher
zaix() {
    local branch_name
    if [ -z "$1" ]; then
        branch_name="worktree-$(date +%Y%m%d-%H%M%S)"
    else
        branch_name="$1"
    fi
    git worktree add "../$branch_name" -b "$branch_name" && \
    cd "../$branch_name" || return 1

    export ANTHROPIC_AUTH_TOKEN="your-zai-api-key"
    export ANTHROPIC_BASE_URL="https://api.z.ai/api/anthropic"
    export API_TIMEOUT_MS="3000000"
    claude --model sonnet --permission-mode plan
}

Windows PowerShell Z.AI + Worktree

# Z.AI + Claude Code worktree launcher
function zaix {
    param([string]$BranchName)
    if (-not $BranchName) {
        $BranchName = "worktree-$(Get-Date -Format 'yyyyMMdd-HHmmss')"
    }
    git worktree add "../$BranchName" -b $BranchName
    if ($LASTEXITCODE -eq 0) {
        Set-Location "../$BranchName"
        $env:ANTHROPIC_AUTH_TOKEN = "your-zai-api-key"
        $env:ANTHROPIC_BASE_URL = "https://api.z.ai/api/anthropic"
        $env:API_TIMEOUT_MS = "3000000"
        claude --model sonnet --permission-mode plan
    }
}

Windows CMD Z.AI + Worktree

Create zaix.bat:

@echo off
setlocal enabledelayedexpansion
if "%~1"=="" (
    for /f "tokens=2 delims==" %%I in ('wmic os get localdatetime /value') do set datetime=%%I
    set branch_name=worktree-!datetime:~0,8!-!datetime:~8,6!
) else (
    set branch_name=%~1
)
git worktree add "../%branch_name%" -b "%branch_name%"
if %errorlevel% equ 0 (
    cd "../%branch_name%"
    set ANTHROPIC_AUTH_TOKEN=your-zai-api-key
    set ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
    set API_TIMEOUT_MS=3000000
    claude --model sonnet --permission-mode plan
)
endlocal

Usage:

# Create worktree with custom name
zaix feature-auth

# Create worktree with auto-generated timestamp name
zaix

6.1.x Claude Code GitHub Actions

Claude Code can be integrated with GitHub Actions to automate AI-powered workflows. With a simple @claude mention in PRs or issues, Claude can analyze code, implement features, fix bugs, and follow project standards defined in CLAUDE.md.

Key capabilities:

Respond to @claude mentions in issues and pull requests
Create and modify code through pull requests
Follow project-specific guidelines from CLAUDE.md
Execute slash commands like /review

Reference: Claude Code GitHub Actions Documentation

Z.AI Integration Example

Create .github/workflows/claude.yml:

Click to expand workflow YAML

name: Claude Code

on:
  issue_comment:
    types: [created]
  pull_request_review_comment:
    types: [created]
  issues:
    types: [opened, assigned]
  pull_request_review:
    types: [submitted]

jobs:
  claude:
    if: |
      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||
      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||
      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||
      (github.event_name == 'issues' && (contains(github.event.issue.body, '@claude') || contains(github.event.issue.title, '@claude')))
    runs-on: ubuntu-latest
    permissions:
      contents: write
      pull-requests: write
      issues: write
      id-token: write
      actions: read
    steps:
      - name: Checkout repository
        uses: actions/checkout@v5
        with:
          fetch-depth: 1

      - name: Run Claude Code
        id: claude
        uses: anthropics/claude-code-action@v1
        env:
          ANTHROPIC_BASE_URL: https://api.z.ai/api/anthropic
          API_TIMEOUT_MS: 3000000
        with:
          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
          claude_args: |
            --model claude-opus
            --max-turns 100

Workflow Explanation

Component	Purpose
Event Triggers	Listens for `issue_comment`, `pull_request_review_comment`, `issues`, and `pull_request_review` events
Conditional (`if`)	Only runs when `@claude` is mentioned in the comment/issue body or title
Permissions	`contents: write` for code changes, `pull-requests: write` for PRs, `issues: write` for issue responses, `actions: read` for CI results
ANTHROPIC_BASE_URL	Routes API calls through Z.AI endpoint for higher quotas
API_TIMEOUT_MS	Extended timeout (50 minutes) for complex operations
claude_args	Uses `claude-opus` model with up to 100 turns for complex tasks

Setup Requirements

Add API key as secret: Go to repository Settings → Secrets and variables → Actions → Add ANTHROPIC_API_KEY with your Z.AI API key
Create workflow file: Save the YAML above to .github/workflows/claude.yml
Usage: Mention @claude in any issue or PR comment to trigger the workflow

7. Cost & Rate Management

7.1 Claude Plan Weekly Rate Limits

From August 28, 2025, weekly rate limits apply in addition to monthly 50x 5hr session limits:

Plan	Sonnet 4 (hrs/week)	Opus 4 (hrs/week)
Pro	40–80	–
Max ($100/mo)	140–280	15–35
Max ($200/mo)	240–480	24–40

7.2 Cost Tracking

Use the Usage Metrics MCP server to track costs:

claude mcp add --transport stdio metrics -s user -- uv run --directory /path/to/mcp-server metrics-server

Or use the /ccusage-daily slash command for detailed cost analysis.

7.3 Optimization Tips

Use Z.AI for higher quotas at lower cost
Configure status lines to monitor real-time usage
Use git worktrees for parallel sessions without duplicating quota
Leverage CoD mode in code-searcher for 80% token reduction

REFERENCE SECTION

8. Configuration Reference

8.1 Settings Scopes

Scope	Location	Affects	Shared?
Managed	System directories	All users	By IT
User	`~/.claude/settings.json`	You (all projects)	No
Project	`.claude/settings.json`	All collaborators	Yes
Local	`.claude/settings.local.json`	You (this project)	No

Precedence (highest to lowest):

Enterprise policies
Command line arguments
Local project settings
Shared project settings
User settings

8.2 settings.json Options

Key	Description	Example
`apiKeyHelper`	Script to generate auth value	`/bin/generate_temp_api_key.sh`
`cleanupPeriodDays`	Days to retain chat transcripts (default: 30)	`20`
`env`	Environment variables for every session	`{"FOO": "bar"}`
`includeCoAuthoredBy`	Include `co-authored-by Claude` in commits	`false`
`permissions`	Permission rules (see below)

Permission Settings

Key	Description	Example
`allow`	Array of allowed tool use rules	`["Bash(git diff:*)"]`
`deny`	Array of denied tool use rules	`["WebFetch", "Bash(curl:*)"]`
`additionalDirectories`	Additional working directories	`["../docs/"]`
`defaultMode`	Default permission mode	`"acceptEdits"`
`disableBypassPermissionsMode`	Prevent bypass mode	`"disable"`

8.3 Environment Variables

Variable	Purpose
`ANTHROPIC_API_KEY`	API key for Claude SDK
`ANTHROPIC_AUTH_TOKEN`	Custom auth header value
`ANTHROPIC_BASE_URL`	Custom API endpoint
`ANTHROPIC_MODEL`	Custom model name
`ANTHROPIC_SMALL_FAST_MODEL`	Haiku-class model for background tasks
`BASH_DEFAULT_TIMEOUT_MS`	Default bash command timeout
`BASH_MAX_TIMEOUT_MS`	Maximum bash command timeout
`BASH_MAX_OUTPUT_LENGTH`	Max characters before truncation
`CLAUDE_CODE_MAX_OUTPUT_TOKENS`	Max output tokens per request
`CLAUDE_CODE_USE_BEDROCK`	Use Amazon Bedrock
`CLAUDE_CODE_USE_VERTEX`	Use Google Vertex AI
`DISABLE_AUTOUPDATER`	Disable automatic updates
`DISABLE_TELEMETRY`	Opt out of Statsig telemetry
`HTTP_PROXY` / `HTTPS_PROXY`	Proxy server settings
`MAX_THINKING_TOKENS`	Force thinking budget
`MCP_TIMEOUT`	MCP server startup timeout
`MCP_TOOL_TIMEOUT`	MCP tool execution timeout
`MAX_MCP_OUTPUT_TOKENS`	Max MCP response tokens (default: 25000)

8.4 File Locations by OS

File	macOS	Linux	Windows
User settings	`~/.claude/settings.json`	`~/.claude/settings.json`	`%USERPROFILE%\.claude\settings.json`
Project settings	`.claude/settings.json`	`.claude/settings.json`	`.claude\settings.json`
Local settings	`.claude/settings.local.json`	`.claude/settings.local.json`	`.claude\settings.local.json`

8.5 Tools Available to Claude

Tool	Description	Permission Required
Agent	Runs sub-agents for complex tasks	No
Bash	Executes shell commands	Yes
Edit	Makes targeted file edits	Yes
Glob	Finds files by pattern	No
Grep	Searches file contents	No
LS	Lists files and directories	No
MultiEdit	Multiple edits atomically	Yes
NotebookEdit	Modifies Jupyter cells	Yes
NotebookRead	Reads Jupyter notebooks	No
Read	Reads file contents	No
TodoRead	Reads task list	No
TodoWrite	Creates/manages task lists	No
WebFetch	Fetches URL content	Yes
WebSearch	Performs web searches	Yes
Write	Creates/overwrites files	Yes

8.6 Configuration Commands

# List settings
claude config list

# Get a setting
claude config get <key>

# Set a setting
claude config set <key> <value>

# Add to a list setting
claude config add <key> <value>

# Remove from a list setting
claude config remove <key> <value>

# Global configuration
claude config set -g <key> <value>

Global Configuration Options

Key	Description	Example
`autoUpdates`	Enable automatic updates (default: true)	`false`
`preferredNotifChannel`	Notification method	`iterm2`, `terminal_bell`, `notifications_disabled`
`theme`	Color theme	`dark`, `light`, `light-daltonized`, `dark-daltonized`
`verbose`	Show full command outputs	`true`

9. Resources

Official Documentation

YouTube Guides

Claude Code with Opus 4.5 - Alex Finn
How To Master Claude Code (Anthropic's Official 7-Hour Course) - Anthropic Official
Claude Code Overview - Matt Maher
VS Code Beginner Guide
Ralph Wiggum Tutorial - Greg Isenberg & Ryan Carson

Other Resources

Advent of Claude: 31 Days - Ado Kukic
40+ Claude Code Tips
OpenClaw: Security, Deployment & Cost Guide - For using Claude LLM models with OpenClaw

FilesExpand file tree

README-v2.md

Latest commit

History

README-v2.md

File metadata and controls

My Claude Code Setup

Table of Contents

Quick Start

Option A: Clone Entire Repository (New Projects)

Option B: Selective Copy (Existing Projects)

Option C: Download Specific Files (GitHub)

BEGINNER SECTION

1. Getting Started

1.1 What This Repository Provides

1.2 Prerequisites

1.3 Installation Steps

1.4 Optional Enhancements

1.5 Useful Learning Resources

2. Memory Bank System

2.1 What is Memory Bank?

2.2 Core Context Files

2.3 Setting Up Memory Bank

2.4 Updating Memory Bank

2.5 Memory Bank Example Output

INTERMEDIATE SECTION

3. Extending Claude Code

3.1 Plugin System

Official Anthropic Marketplace

Recommended Official Plugins

Third-Party Marketplaces

3.2 MCP Servers

Recommended MCP Servers

Installation Commands

Verify MCP Servers

Usage Metrics Example

4. Development Workflows

4.1 Git Worktrees for Parallel Sessions

macOS / Linux Shell Functions

Usage

Worktree Management

Environment Setup

The .worktreeinclude File

Claude Desktop Worktree Location

Local Ignores Without .gitignore

4.2 Status Lines

Example Status Line Script

4.3 IDE Integration

ADVANCED SECTION

5. Customization

5.1 Custom Subagents

Included Subagents

memory-bank-synchronizer

code-searcher

get-current-datetime

ux-design-expert

zai-cli

codex-cli

5.2 Custom Skills

Included Skills

5.3 Custom Slash Commands

/anthropic Commands

/ccusage Commands

/cleanup Commands

/documentation Commands

/security Commands

/architecture Commands

/promptengineering Commands

/refactor Commands

5.4 Hooks System

Included Hooks

6. Alternative Providers

6.1 Using Z.AI with Claude Code

Overview

Pricing & Plans

Privacy & Data Handling

Prerequisites

Setup Instructions

Shell Function Launchers

Model Mapping

`/anthropic` Commands

`/ccusage` Commands

`/cleanup` Commands

`/documentation` Commands

`/security` Commands

`/architecture` Commands

`/promptengineering` Commands

`/refactor` Commands