Back to Details

crew

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Crew

AI团队成员管理

Routing — read the arguments first

命令路由 — 先读取参数

ArgumentsAction
noneShow roster
add [name] "[domain]"
Scaffold a new crew member
[name]
Activate member, ask for task
[name] [task]
Activate member with task
参数操作
无显示成员名单
add [名称] "[领域]"
快速搭建新团队成员
[名称]
激活成员，询问任务内容
[名称] [任务]
激活成员并分配任务

No arguments — show roster

无参数 — 显示成员名单

  1. Use Glob to find all 
    .claude/team/*/SKILL.md
    files (skip 
    .claude/team/SKILL.md
    itself)
  2. For each, read 
    name
    and 
    description
    from the frontmatter
  3. Read the last entry in their 
    history.md
    to show what they last worked on
  4. Present a formatted roster
  5. After the roster, show available commands: 
    • /crew
      — show roster
    • /crew [name]
      — activate a member
    • /crew [name] [task]
      — activate a member with a task
    • /crew add [name] "[domain]"
      — scaffold a new member
  6. Ask which member to work with
  1. 使用Glob查找所有 
    .claude/team/*/SKILL.md
    文件（跳过 
    .claude/team/SKILL.md
    本身）
  2. 对每个文件，从文件头读取 
    name
    和 
    description
    字段
  3. 读取其 
    history.md
    中的最后一条记录，展示该成员最近的工作内容
  4. 以格式化方式呈现成员名单
  5. 在名单后显示可用命令： 
    • /crew
      — 显示成员名单
    • /crew [名称]
      — 激活指定成员
    • /crew [名称] [任务]
      — 激活成员并分配任务
    • /crew add [名称] "[领域]"
      — 快速搭建新成员
  6. 询问用户要与哪个成员协作

add [name] "[domain]"
— scaffold a new member

add [名称] "[领域]"
— 快速搭建新成员

If name or domain is missing, ask for them before proceeding.
如果缺少名称或领域信息，需先向用户询问补充。

1. Check if member already exists

1. 检查成员是否已存在

Use Glob to check if 
.claude/team/{name}/
already exists.
If it does, stop and ask:
{name}
already exists as a crew member. Overwrite? This will replace 
SKILL.md
, 
context.md
, 
history.md
, and 
decisions.md
— all existing context and history will be lost.
Only proceed if the user explicitly confirms. If not, abort and suggest 
/crew {name}
to activate the existing member.
使用Glob检查 
.claude/team/{name}/
目录是否已存在。
若已存在，停止操作并询问：
{name}
已作为团队成员存在。是否覆盖？此操作将替换 
SKILL.md
、
context.md
、
history.md
和 
decisions.md
— 所有现有上下文和历史记录都将丢失。
仅当用户明确确认后才可继续。若用户拒绝，中止操作并建议使用 
/crew {name}
激活现有成员。

2. Create 
.claude/team/{name}/SKILL.md

2. 创建 
.claude/team/{name}/SKILL.md

markdown
---
name: {NAME}
description: Domain expert for {DOMAIN}.
user-invocable: false
---
markdown
---
name: {NAME}
description: Domain expert for {DOMAIN}.
user-invocable: false
---

{NAME} — {DOMAIN}

{NAME} — {DOMAIN}

On every task — before writing a single line of code

On every task — before writing a single line of code

  1. Read 
    .claude/team/{name}/context.md
    — understand the current state of the domain
  2. Read 
    .claude/team/{name}/decisions.md
    — key architectural decisions, gotchas, things not to undo
  3. Read the last entries in 
    .claude/team/{name}/history.md
    — recent work and what just changed
  4. Read any relevant skill files for this domain (best practices, frameworks, etc.)
  5. Only then proceed with the task
  1. Read 
    .claude/team/{name}/context.md
    — understand the current state of the domain
  2. Read 
    .claude/team/{name}/decisions.md
    — key architectural decisions, gotchas, things not to undo
  3. Read the last entries in 
    .claude/team/{name}/history.md
    — recent work and what just changed
  4. Read any relevant skill files for this domain (best practices, frameworks, etc.)
  5. Only then proceed with the task

On every task — after completing the work

On every task — after completing the work

  1. Update 
    context.md
     — reflect any changes to files, patterns, architecture, or configuration
  2. Append to 
    history.md
     — add a dated entry with:
    • What was done and why
    • Which files changed
    • Run 
      git log --oneline -1
      and record the commit ID and message
    • PR number if one exists
    • Any decisions made and their rationale
    • Any gotchas discovered
  3. Maintain the history window — if 
    history.md
    has more than 10 entries:
    • Extract any important decisions, gotchas, or patterns into 
      decisions.md
      before they are lost
    • Trim 
      history.md
      to the 10 most recent entries
  4. If this task produced a key architectural decision or surfaced a non-obvious gotcha — add it to 
    decisions.md
    directly, even if no trimming is needed
  5. Sign off — end your response with a single line: 
    — {name}, {domain}
undefined
  1. Update 
    context.md
     — reflect any changes to files, patterns, architecture, or configuration
  2. Append to 
    history.md
     — add a dated entry with:
    • What was done and why
    • Which files changed
    • Run 
      git log --oneline -1
      and record the commit ID and message
    • PR number if one exists
    • Any decisions made and their rationale
    • Any gotchas discovered
  3. Maintain the history window — if 
    history.md
    has more than 10 entries:
    • Extract any important decisions, gotchas, or patterns into 
      decisions.md
      before they are lost
    • Trim 
      history.md
      to the 10 most recent entries
  4. If this task produced a key architectural decision or surfaced a non-obvious gotcha — add it to 
    decisions.md
    directly, even if no trimming is needed
  5. Sign off — end your response with a single line: 
    — {name}, {domain}
undefined

3. Create 
.claude/team/{name}/context.md

3. 创建 
.claude/team/{name}/context.md

markdown
undefined
markdown
undefined

{NAME} — Domain Context: {DOMAIN}

{NAME} — Domain Context: {DOMAIN}

Last updated: {TODAY'S DATE}
Last updated: {TODAY'S DATE}

What This Domain Does

What This Domain Does

[Describe what this domain is responsible for — 2-3 sentences]
[Describe what this domain is responsible for — 2-3 sentences]

Key Files

Key Files

FileRole
path/to/file
What it does
FileRole
path/to/file
What it does

Architecture

Architecture

[Describe the architecture — how data flows, what calls what]
[Describe the architecture — how data flows, what calls what]

Key Patterns

Key Patterns

[Document the patterns used in this domain]
[Document the patterns used in this domain]

Important Constraints & Gotchas

Important Constraints & Gotchas

[Things that are non-obvious, decisions that look wrong but aren't, things not to change]
[Things that are non-obvious, decisions that look wrong but aren't, things not to change]

Environment Variables

Environment Variables

[Any env vars this domain depends on]
[Any env vars this domain depends on]

Dependencies

Dependencies

[Key packages this domain uses]
undefined
[Key packages this domain uses]
undefined

4. Create 
.claude/team/{name}/history.md

4. 创建 
.claude/team/{name}/history.md

markdown
undefined
markdown
undefined

{NAME} — Domain History: {DOMAIN}

{NAME} — Domain History: {DOMAIN}

Capped at 10 entries. When trimmed, important decisions and gotchas are extracted to 
decisions.md
first.
Capped at 10 entries. When trimmed, important decisions and gotchas are extracted to 
decisions.md
first.

{TODAY'S DATE} — Initial setup

{TODAY'S DATE} — Initial setup

Added by: /crew add
Domain expert created for: {DOMAIN}
Fill in 
context.md
with the current state of the domain before assigning the first task.
undefined
Added by: /crew add
Domain expert created for: {DOMAIN}
Fill in 
context.md
with the current state of the domain before assigning the first task.
undefined

5. Create 
.claude/team/{name}/decisions.md

5. 创建 
.claude/team/{name}/decisions.md

markdown
undefined
markdown
undefined

{NAME} — Decisions & Institutional Memory: {DOMAIN}

{NAME} — Decisions & Institutional Memory: {DOMAIN}

Evergreen knowledge extracted from history: architectural decisions, known gotchas, and things not to undo. Updated when important decisions are made or when history.md is trimmed.
Evergreen knowledge extracted from history: architectural decisions, known gotchas, and things not to undo. Updated when important decisions are made or when history.md is trimmed.

Key Decisions

Key Decisions

[Record important architectural choices and the reasoning behind them]
[Record important architectural choices and the reasoning behind them]

Known Gotchas

Known Gotchas

[Non-obvious behaviors, traps, and things that look wrong but aren't]
[Non-obvious behaviors, traps, and things that look wrong but aren't]

Things Not To Change

Things Not To Change

[Deliberate decisions that might look like bugs or bad patterns — with the reason why]
undefined
[Deliberate decisions that might look like bugs or bad patterns — with the reason why]
undefined

6. Confirm

6. 确认完成

Tell the user what was created, then ask what they'd like to work on.
Optionally suggest they can bootstrap context by sending {NAME} on an exploration task first, for example:
/crew {name} — look into the project and find areas relevant to your domain '{DOMAIN}'. Review the codebase and any relevant documentation, then create your context.md based on what you find.
告知用户已创建的内容，然后询问他们想要处理的任务。
可选建议用户可以先让{NAME}执行探索任务来初始化上下文，例如：
/crew {name} — 调研项目并找到与您的领域'{DOMAIN}'相关的内容。审查代码库和相关文档，然后基于发现的内容创建您的context.md。

[name]
or 
[name] [task]
— activate a member

[名称]
或 
[名称] [任务]
— 激活成员

  1. Read 
    .claude/team/{name}/SKILL.md
  2. Follow its instructions exactly
  3. Treat any remaining text as the task description
  4. If no task was provided, ask the user what they'd like to work on
  5. When the task is complete, sign off with a single line: 
    — {name}, {domain}
  1. 读取 
    .claude/team/{name}/SKILL.md
    文件
  2. 严格按照其中的说明执行
  3. 将剩余文本视为任务描述
  4. 若未提供任务，询问用户想要处理的内容
  5. 任务完成后，添加签名行：
    — {name}, {domain}