= Agents · 
= Commands · 
= Skills
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ 
→ <img src="What is this?
A curated reference and knowledge base documenting Claude Code's features and best practices, covering agents, commands, skills, workflows, hooks, MCP servers, and more, organized as linked tables rather than a single tutorial.
A curated reference and knowledge base documenting Claude Code's features and best practices, covering agents, commands, skills, workflows, hooks, MCP servers, and more, organized as linked tables rather than a single tutorial.Is this a tool I install, or just a reference guide?It is a reference repository, not installable software: a big table of Claude Code concepts (Subagents, Commands, Skills, Hooks, MCP Servers, Plugins) with links to docs, blogs and related skill marketplaces.
It is a reference repository, not installable software: a big table of Claude Code concepts (Subagents, Commands, Skills, Hooks, MCP Servers, Plugins) with links to docs, blogs and related skill marketplaces.How do I actually use this repo?Open the README and work through the CONCEPTS and DEVELOPMENT WORKFLOWS tables; each row links to the relevant Claude Code feature or external doc, and you copy patterns like .claude/agents/*.md, .claude/commands/*.md, or .claude/skills/*/SKILL.md into your own project.
Open the README and work through the CONCEPTS and DEVELOPMENT WORKFLOWS tables; each row links to the relevant Claude Code feature or external doc, and you copy patterns like .claude/agents/*.md, .claude/commands/*.md, or .claude/skills/*/SKILL.md into your own project.Do I need Claude Code itself to make use of this?Yes, everything documented here (agents, commands, skills, hooks, settings) is a feature of Anthropic's Claude Code CLI, so you need Claude Code installed to apply any of it.
Yes, everything documented here (agents, commands, skills, hooks, settings) is a feature of Anthropic's Claude Code CLI, so you need Claude Code installed to apply any of it.Is this an official Anthropic resource?No, it reads as an independently curated cheat sheet rather than official Anthropic documentation, so 'Hot' entries like Auto Mode, Agent Teams, or Scheduled Tasks may be newer or change quickly as Claude Code evolves.
No, it reads as an independently curated cheat sheet rather than official Anthropic documentation, so 'Hot' entries like Auto Mode, Agent Teams, or Scheduled Tasks may be newer or change quickly as Claude Code evolves.What is the Research to Ship pattern mentioned in the repo?It is the common architectural pattern the author says all major Claude Code workflows converge on, tying together the Command to Agent to Skill chain described in the orchestration-workflow section.
It is the common architectural pattern the author says all major Claude Code workflows converge on, tying together the Command to Agent to Skill chain described in the orchestration-workflow section.Which third-party skill collections does it point to?It lists star-ranked community skill and plugin collections such as Superpowers, Matt Pocock Skills, Everything Claude Code, Spec Kit, gstack, and agent-skills, along with counts of their agents, commands, and skills.
It lists star-ranked community skill and plugin collections such as Superpowers, Matt Pocock Skills, Everything Claude Code, Spec Kit, gstack, and agent-skills, along with counts of their agents, commands, and skills.Who is Boris Cherny, mentioned at the top?He is referenced via a linked X profile near the top of the README as a notable figure tied to Claude Code, with the repo opening on that callout before diving into the feature tables.
He is referenced via a linked X profile near the top of the README as a notable figure tied to Claude Code, with the repo opening on that callout before diving into the feature tables.What counts as a 'Hot' feature in the list?The Hot table highlights newer or trending Claude Code capabilities like Ultrareview, Auto Mode, Fast Mode, Agent Teams, and Scheduled Tasks, each paired with its trigger command or flag.
The Hot table highlights newer or trending Claude Code capabilities like Ultrareview, Auto Mode, Fast Mode, Agent Teams, and Scheduled Tasks, each paired with its trigger command or flag.What are similar projects?claude-code-ultimate-guide, claude-code-tips, context-engineering-intro, harness-engineering, Agent-Learning-Hub, vibe-coding-cn, ai-system-design, awesome-claude-skills
claude-code-ultimate-guide, claude-code-tips, context-engineering-intro, harness-engineering, Agent-Learning-Hub, vibe-coding-cn, ai-system-design, awesome-claude-skillsWhich topics is it in?agentic-engineering, claude-ai, claude-code-skills, agentic-workflow, agentic-coding, vibe-coding, cheatsheet, context-engineering, best-practices, knowledge-base, prompt-engineering, agentic-ai
agentic-engineering, claude-ai, claude-code-skills, agentic-workflow, agentic-coding, vibe-coding, cheatsheet, context-engineering, best-practices, knowledge-base, prompt-engineering, agentic-ai