luongnv89/claude-howto

Claude How To is a structured, example-driven learning guide for Claude Code, Anthropic's AI-powered command-line coding assistant. It solves the problem that the official Claude Code documentation describes individual features but does not show how to combine them into real workflows, leaving users without a clear learning path from basic usage to advanced capabilities. The guide is organized into ten tutorial modules covering slash commands, memory management, checkpoints, CLI usage, skills, hooks, MCP (Model Context Protocol) servers, subagents, advanced features, and plugins. Each module includes visual diagrams showing how the feature works internally, copy-paste configuration templates you can immediately drop into your project, and self-assessment quizzes accessible directly within Claude Code by typing slash commands like /self-assessment or /lesson-quiz hooks. The learning path is designed to be progressive: earlier modules build foundational knowledge that later modules rely on. The guide estimates the full path takes eleven to thirteen hours and includes time estimates per module so learners can plan accordingly. Templates cover CLAUDE.md project memory files, hook scripts, subagent definitions, skill configurations, and full plugin bundles, giving concrete starting points rather than just explanations. You would use this when you want to move beyond basic Claude Code prompting and learn how to build automated workflows using hooks, delegate tasks to subagents, create reusable skills, integrate external tools via MCP, or combine multiple features into end-to-end pipelines like automated code reviews or deployment processes. The repository's primary language is Python for its example scripts, but the guide content covers configuration in JSON, YAML, and Markdown. It is synced with Claude Code releases and is available in English, Vietnamese, Chinese, Ukrainian, and Japanese.