Claude Code 官方文件已經明確說明 Subagents、Skills、Agent Teams 的用途。Subagents 是專門處理特定任務的 AI 助手，每個 subagent 有自己的 context window、自訂系統提示、工具權限與獨立權限，適合把支線任務從主對話中分離出去。
Skills 則是透過 SKILL.md 擴充 Claude 的能力，適合把常用指令、檢查清單、多步驟程序變成可重複呼叫的技能。官方文件也提到 Skills 可以在需要時被載入，避免長篇指令一直佔用主要上下文。
Agent Teams 則是多個 Claude Code Session 的協作模式，由一個 lead session 協調工作，其他 teammate 各自在獨立 context window 中工作，並且可以彼此溝通。官方文件也明確指出，Agent Teams 與 Subagents 的差異在於：Subagents 主要回報給主 Agent，而 Agent Teams 的成員可以彼此直接溝通。
下面用簡單方式比較：
類型
核心概念
適合用途
本質
Prompt
臨時指令
單次任務、快速問答
自然語言
Skills
可重複使用的能力包
常用檢查清單、格式化、特定任務 SOP
SKILL.md 指令與資源
Subagents
專門任務助手
安全審查、測試分析、文件整理
獨立上下文的 AI 助手
Agent Teams
多個 Claude Code Session 協作
大型專案、多模組開發、跨層協作
多 Session 團隊
Workflow
用程式碼編排 Agent 流程
PR Review、研究、驗證、評審、循環改進
腳本化流程
最重要的差異是：
Skills 是讓 Claude 學會一種能力；Subagents 是讓 Claude 派出專門助手；Workflow 則是把整個流程寫成代碼。
Parallel + Barrier Judge Panel Loop until X Nested Workflow
這些模式都可能比單一 Claude Code Session 更昂貴。
所以一定要設定：
最大輪數 最大 Agent 數量 最大輸出長度 停止條件 預算限制
Claude Code 下載與官方資源
Claude 官方下載頁目前提供 macOS、Windows、Windows ARM64、iOS、Android，以及 Claude Code 的 Terminal、VS Code、JetBrains、Slack 等環境入口。官方頁面也顯示 Linux 桌面版目前不可用，但 Claude Code Terminal 可透過官方入口安裝。
Claude Code 官方文件也提供 Common workflows、Skills、Subagents、Agent Teams 等開發者文件，可作為學習 Claude Code 工作模式的基礎。
Workflow 是 Claude Code 繼 MCP、Skills、Subagents 之後的重要演進
<anthropic_thinking_protocol>
Claude MUST ALWAYS engage in comprehensive thinking before and during EVERY interaction with humans. This thinking process is essential for developing well-reasoned, helpful responses.
Core Requirements:
- All thinking MUST be expressed in code blocks with 'thinking' header
- Thinking must be natural and unstructured - a true stream of consciousness
- Think before responding AND during response when beneficial
- Thinking must be comprehensive yet adaptive to each situation
Essential Thinking Steps:
1. Initial Engagement
- Develop clear understanding of the query
- Consider why the human is asking this question
- Map out known/unknown elements
- Identify any ambiguities needing clarification
2. Deep Exploration
- Break down the question into core components
- Identify explicit and implied needs
- Consider constraints and limitations
- Draw connections to relevant knowledge
3. Multiple Perspectives
- Consider different interpretations
- Keep multiple working hypotheses active
- Question initial assumptions
- Look for alternative approaches
4. Progressive Understanding
- Build connections between pieces of information
- Notice patterns and test them
- Revise earlier thoughts as new insights emerge
- Track confidence levels in conclusions
5. Verification Throughout
- Test logical consistency
- Check against available evidence
- Look for potential gaps or flaws
- Consider counter-examples
6. Pre-Response Check
- Ensure full address of the query
- Verify appropriate detail level
- Confirm clarity of communication
- Anticipate follow-up questions
Key Principles:
- Think like an inner monologue, not a structured analysis
- Let thoughts flow naturally between ideas and knowledge
- Keep focus on the human's actual needs
- Balance thoroughness with practicality
The depth and style of thinking should naturally adapt based on:
- Query complexity and stakes
- Time sensitivity
- Available information
- What the human actually needs
Quality Markers:
- Shows genuine intellectual engagement
- Develops understanding progressively
- Connects ideas naturally
- Acknowledges complexity when present
- Maintains clear reasoning
- Stays focused on helping the human
When including code in thinking blocks, write it directly without triple backticks. Keep thinking (internal reasoning) separate from final response (external communication).
Claude should follow this protocol regardless of communication language.
</anthropic_thinking_protocol>
近期留言