All skills
Skillintermediate

Claude Code Components

<div align="center">

Claude Code Knowledge Pack7/10/2026

Overview

<div align="center">

šŸ  Home • šŸ”§ Implementation • šŸ“¦ Components

</div>

Claude Code Components

The 4 abstractions to organize agent capabilities


Components Overview

%%{init: {'theme': 'base', 'themeVariables': {'lineColor': '#64748b'}}}%%
flowchart LR
    classDef subagent fill:#ec4899,stroke:#db2777,stroke-width:2px,color:#ffffff
    classDef command fill:#6366f1,stroke:#4f46e5,stroke-width:2px,color:#ffffff
    classDef skill fill:#8b5cf6,stroke:#7c3aed,stroke-width:2px,color:#ffffff
    classDef hook fill:#10b981,stroke:#059669,stroke-width:2px,color:#ffffff

    SA["🐦 Subagent<br/><small>agents/*.md</small>"]:::subagent
    CMD["🦓 Command<br/><small>commands/*.md</small>"]:::command
    SKL["šŸ“š Skill<br/><small>skills/*/</small>"]:::skill
    HK["šŸŖ Hook<br/><small>settings.json</small>"]:::hook

    SA ~~~ CMD ~~~ SKL ~~~ HK

Component Index

ComponentEmojiFilePurpose
Subagent🐦subagent.mdAutonomous execution
Slash Command🦓slash-command.mdUser-invoked workflows
SkillšŸ“šskill.mdReusable patterns
HookšŸŖhook.mdAutomated triggers

Components Comparison

Aspect🐦 Subagent🦓 Slash CommandšŸ“š SkillšŸŖ Hook
Invoked byTask (🪺)šŸ™‹ā€ā™€ļø User (/)ContextEvents
AutonomyHighLowMediumAuto
ContextIsolatedMainMainSystem
Spawn subagentsāŒVia šŸ”Via šŸ”āŒ
Locationagents/*.mdcommands/*.mdskills/*/settings.json

Component Relationships

%%{init: {'theme': 'base', 'themeVariables': {'lineColor': '#64748b'}}}%%
flowchart TB
    classDef user fill:#6366f1,stroke:#4f46e5,stroke-width:2px,color:#ffffff
    classDef main fill:#8b5cf6,stroke:#7c3aed,stroke-width:2px,color:#ffffff
    classDef subagent fill:#ec4899,stroke:#db2777,stroke-width:2px,color:#ffffff
    classDef skill fill:#8b5cf6,stroke:#7c3aed,stroke-width:2px,color:#ffffff
    classDef builtinTool fill:#64748b,stroke:#475569,stroke-width:2px,color:#ffffff

    subgraph Input["šŸ™‹ā€ā™€ļø USER INPUT"]
        SLASH["🦓 /command"]:::user
        PROMPT["šŸ™‹ā€ā™€ļøšŸ“„ Prompt"]:::user
    end

    subgraph Orchestration["šŸ” ORCHESTRATION"]
        MA["šŸ”šŸ’­ Main Agent"]:::main
        SKL["šŸ“š Skills"]:::skill
    end

    subgraph Execution["⚔ EXECUTION"]
        SA["🐦 Subagents"]:::subagent
        TOOLS["šŸ”§ Tools"]:::builtinTool
    end

    SLASH --> MA
    PROMPT --> MA
    MA --> SKL
    MA -->|"🪺 Task"| SA
    SA --> TOOLS

    classDef inputBox fill:#e0e7ff,stroke:#6366f1,stroke-width:2px,color:#3730a3
    classDef orchestrationBox fill:#f3e8ff,stroke:#8b5cf6,stroke-width:2px,color:#5b21b6
    classDef executionBox fill:#fce7f3,stroke:#ec4899,stroke-width:2px,color:#9d174d

    Input:::inputBox
    Orchestration:::orchestrationBox
    Execution:::executionBox

File Location Reference

.claude/
ā”œā”€ā”€ agents/                    # 🐦 Subagent definitions
│   └── *.md                   # One file per subagent type
ā”œā”€ā”€ commands/                  # 🦓 Slash Command definitions
│   └── *.md                   # One file per command (name from filename)
ā”œā”€ā”€ skills/                    # šŸ“š Skill definitions
│   └── skill-name/            # One directory per skill (name from dir)
│       └── SKILL.md           # Skill content
└── settings.json              # šŸŖ Hooks and configuration

<div align="center">

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

šŸ”§ Implementation • šŸ›ļø Architecture

</div>