Guides & workflows
Long-form source material, selected for substantive coverage and paired with editorial notes. Upstream publication dates and behavior may differ from the current product.
Core Concepts
8 guidesReference for Claude Code startup flags, top level subcommands, and startup environment variables when launching Claude Code from the terminal. Table of Contents 1. Session Management 2. Mod…
Claude Code commands — frontmatter fields and official built in slash commands. Frontmatter Fields (20) Field Type Required Description name string No Command name shown in the / menu. Defau…
MCP (Model Context Protocol) servers extend Claude Code with connections to external tools, databases, and APIs. This guide covers recommended servers for daily use and configuration best pr…
Persistent context via CLAUDE.md files — how to write them and how they load in monorepos. 1. Writing a Good CLAUDE.md A well structured CLAUDE.md is the single most impactful way to improve…
Interactive lessons teaching Claude Code features with animated demos. Each power up teaches one thing Claude Code can do that most people miss. Introduced in v2.1.90. Usage Power ups (10) P…
A comprehensive guide to all available configuration options in Claude Code's settings.json files. As of v2.1.296, Claude Code exposes 244+ settings and 396+ environment variables (use the "…
Claude Code skills — frontmatter fields and official bundled skills. Frontmatter Fields (20) Field Type Required Description name string No Display name and /slash command identifier. Defaul…
Claude Code subagents — frontmatter fields and official built in agent types. Frontmatter Fields (16) Field Type Required Description name string Yes Unique identifier using lowercase letter…
Implementation
6 guidesAgent Teams spawn multiple independent Claude Code sessions that coordinate via a shared task list. Unlike subagents (isolated context forks within one session), each teammate gets its own f…
The weather orchestrator command is implemented in this repo as the entry point of the Command → Agent → Skill architecture pattern, demonstrating how commands orchestrate multi step workflo…
/goal keeps your agent working across turns until a condition is satisfied — Claude Code, Codex, and Hermes Agent all support it. The community is converging on a few high leverage prompting…
The /loop skill is used to schedule recurring tasks on a cron interval. Below is a demo of /loop 1m "tell current time" — a simple recurring task that fires every minute. Loop Demo 1. Schedu…
Two skills are implemented in this repo as part of the Command → Agent → Skill architecture pattern, demonstrating two distinct skill invocation patterns: agent skills (preloaded) and skills…
The weather agent is implemented in this repo as an example of the Command → Agent → Skill architecture pattern, demonstrating two distinct skill patterns. Weather Agent File : .claude/agent…
Deep Reports
12 guidesAPI level features (now GA) that reduce token consumption, latency, and improve tool accuracy. Released with Opus/Sonnet 4.6. Table of Contents 1. Overview 2. Programmatic Tool Calling (PTC)…
A comparison of the three extension mechanisms in Claude Code: subagents, commands, and skills. At a Glance Agent Command Skill Location .claude/agents/.md .claude/commands/.md .claude/skill…
Persistent memory for subagents — enabling agents to learn, remember, and build knowledge across sessions. Overview Introduced in Claude Code v2.1.33 (February 2026), the memory frontmatter …
Executive Summary When sending the same message (e.g., "What is the capital of Norway?") through the Claude Agent SDK versus the Claude CLI (Claude Code) , the system prompts accompanying th…
A comprehensive comparison of which Claude Code features are global only ( /.claude/ ) versus which have both global and project level ( .claude/ ) equivalents. Table of Contents 1. Overview…
Executive Summary Based on extensive research, I've analyzed the two tools from your screenshots plus a third major contender. Here's my comprehensive breakdown to help you choose the best o…
When working with Claude Code in a monorepo, understanding how skills are discovered and loaded into context is crucial for organizing your project specific capabilities effectively. Importa…
The rotating words and tips shown under Claude Code's spinner. Extracted from /.local/share/claude/versions/2.1.121 . Spinner words Verb Verb Verb Verb : : : : 1 Accomplishing 48 Discombobul…
Understanding how usage limits work in Claude Code and how to keep working when you hit them. Overview Claude Code on subscription plans (Pro, Max 5x, Max 20x) has usage limits that reset on…
← Back to README Overview Redesign all slides from slide 7 onward around a single running example: the weather reporter agent . The narrative arc matches the TOC visible order (Agents → Skil…
Can a deployed LLM's performance change day to day even though the model weights are frozen? A deep dive into proven causes, infrastructure bugs, and psychological factors. 🔥 Claude Code Op…
Why Claude Code's features are not "just prompts in disguise" — and why the harness is what actually separates toy output from production grade engineering work. Executive Summary A common r…
Tutorials
4 guidesThis guide walks you through installing Claude Code on your machine and authenticating so you can start using it. Step 1: Install Claude Code Choose your operating system: OS Guide Windows w…
Back to Day 0 Prerequisites You need Node.js v18 or higher and npm . Step 1: Install Node.js Option A: Via nodejs.org Download Page with fnm (Recommended) fnm (Fast Node Manager) is official…
Back to Day 0 Node.js Go to nodejs.org Click the "Download Node.js (LTS)" button — this downloads the .msi installer Run the .msi file and click Next through the wizard Accept the defaults, …
Back to Day 0 (Setup) You've installed Claude Code. Now what? This guide walks you through three levels of using it — each one gives you more control over how Claude does its work. Think of …