bin/CLAUDE.md

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Repository Overview

This is a personal utility bin directory containing command-line tools and scripts for development, git operations, and system automation. The scripts are primarily written in bash, Ruby, and Perl, with some Python utilities.

Script Categories and Architecture

Git Utilities

• Git workflow enhancement scripts (git-update, git-remove-merged-branches, git-ai-message, etc.)
• Follow the pattern of accepting remote/branch parameters with sensible defaults
• Include error handling with set -e in bash scripts
• Support dry-run modes where applicable (use -n flag)

Development Tools

• Web development utilities (serve, make-bookmarklet, sri-integrity)
• Image processing tools (scale-app-icons, generate-xcode-imageset, retina-scale)
• Data processing scripts (colours.rb for color conversion, progress for piped data)

System Scripts

• Shell enhancement utilities (hist.rb for command analysis, enable-sudo-touch-id)
• File management tools (find-unused-images, count-chars)

Common Patterns

Script Structure

• All executable scripts start with appropriate shebang (#!/bin/bash, #!/usr/bin/env ruby, etc.)
• Bash scripts use set -e for error handling
• Ruby scripts often process STDIN/command line arguments with explicit argument parsing
• Support both piped input and file arguments where relevant

Argument Handling

• Use "${1:-default}" pattern for optional arguments with defaults in bash
• Ruby scripts use ARGV.shift pattern for argument processing
• Include help/usage when arguments are malformed

Git Script Conventions

• Accept remote name as first argument (default to origin or auto-detect)
• Accept branch name as second argument (default to current branch or master/main)
• Preserve original branch state (stash/unstash, checkout back to original branch)
• Use git rev-parse --abbrev-ref HEAD to get current branch

Data Processing

• Ruby scripts that process data streams use proper buffering and error handling
• Support both byte and line counting modes where applicable
• Use STDERR for progress/status output, STDOUT for data

Development Workflow

This repository has no build system, package.json, or formal testing framework. Scripts are standalone utilities meant to be:

• Executable from anywhere in the system PATH
• Self-contained with minimal dependencies
• Robust with proper error handling