Claude Code Workflow Cheatsheet — The Designer-Polished Edition

Yesterday I posted a Claude Code cheatsheet made entirely with AI graphics. It went viral. Today's the designer-polished version — cleaner, more readable, worth saving again.

What: Claude Code Is a 4-Layer System

I spent the last few weeks living inside Claude Code. Not just using it. Building with it. Breaking it. Figuring out what actually works vs. what wastes your time.

The biggest lesson? Claude Code isn't just a terminal chatbot. It's a 4-layer system:

CLAUDE.md — your project's persistent memory
Skills — knowledge packs Claude auto-invokes when relevant
Hooks — deterministic safety gates (100% enforced, not "suggestions")
Agents — subagents with their own context windows

Most engineers install it, type a prompt, and wonder why the output is mid.

Why: The Gap Between Average and Exceptional

The gap between average and exceptional results comes down to setup:

Run /init on day one to generate your CLAUDE.md
Structure your .claude/ directory with skills, hooks, and permissions
Write descriptions that actually trigger the right skills at the right time
Use the memory hierarchy (global → project → subfolder) to scope context
Set up hooks for safety rules — CLAUDE.md rules are ~70% followed, hooks are 100%

If you're using Claude Code without setting up CLAUDE.md and skills — you're leaving 80% of its power on the table.

How: Everything in One Cheatsheet

I put all of this into a single-page A4 cheatsheet. 14 sections. 3 columns. Everything from first install to the 4-layer architecture.

What's inside:

Getting Started — install → /init → first session
Understanding CLAUDE.md — the WHAT / WHY / HOW framework for writing an effective project memory file
Project File Structure — exact directory layout for a fully configured .claude/ project
The 4-Layer Architecture — CLAUDE.md, Skills, Hooks, Agents — and how they interact
Adding Skills — how to write descriptions that actually trigger correctly
Setting Up Hooks — configuring permissions in settings.json for deterministic safety
Memory File Hierarchy — global → project → subfolder scoping
Daily Workflow Pattern — Plan Mode → Auto-Accept → commit loop (Shift + Tab to toggle)
Quick Reference — keyboard shortcuts, slash commands, and the most useful flags

Save it. Print it. Pin it next to your terminal.

P/S: Credit to Brij Kishore Pandey for the original cheatsheet design. If you want the full breakdown of each section, drop a comment — I might turn this into a series.