command-creator
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseCommand Creator
命令创建器
This skill guides the creation of Claude Code slash commands - reusable workflows that can be invoked with in Claude Code conversations.
/command-name此技能指导你创建Claude Code斜杠命令——这些是可复用的工作流程,在Claude Code对话中可通过调用。
/command-nameAbout Slash Commands
关于斜杠命令
Slash commands are markdown files stored in (project-level) or (global/user-level) that get expanded into prompts when invoked. They're ideal for:
.claude/commands/~/.claude/commands/- Repetitive workflows (code review, PR submission, CI fixing)
- Multi-step processes that need consistency
- Agent delegation patterns
- Project-specific automation
斜杠命令是存储在(项目级)或(全局/用户级)的Markdown文件,调用时会展开为提示词。它们非常适合:
.claude/commands/~/.claude/commands/- 重复性工作流(代码审查、PR提交、CI修复)
- 需要一致性的多步骤流程
- Agent委托模式
- 项目特定的自动化
When to Use This Skill
何时使用此技能
Invoke this skill when users:
- Ask to "create a command" or "make a slash command"
- Want to automate a repetitive workflow
- Need to document a consistent process for reuse
- Say "I keep doing X, can we make a command for it?"
- Want to create project-specific or global commands
当用户出现以下情况时,调用此技能:
- 要求“创建命令”或“制作斜杠命令”
- 希望自动化重复性工作流
- 需要将一致的流程文档化以便复用
- 说“我一直在做X,我们能为此制作一个命令吗?”
- 想要创建项目特定或全局命令
Bundled Resources
附带资源
This skill includes reference documentation for detailed guidance:
- references/patterns.md - Command patterns (workflow automation, iterative fixing, agent delegation, simple execution)
- references/examples.md - Real command examples with full source (submit-stack, ensure-ci, create-implementation-plan)
- references/best-practices.md - Quality checklist, common pitfalls, writing guidelines, template structure
Load these references as needed when creating commands to understand patterns, see examples, or ensure quality.
此技能包含参考文档,供你获取详细指导:
- references/patterns.md - 命令模式(工作流自动化、迭代修复、Agent委托、简单执行)
- references/examples.md - 完整源码的真实命令示例(submit-stack、ensure-ci、create-implementation-plan)
- references/best-practices.md - 质量检查表、常见陷阱、编写指南、模板结构
创建命令时,可根据需要加载这些参考文档,以了解模式、查看示例或确保质量。
Command Structure Overview
命令结构概述
Every slash command is a markdown file with:
markdown
---
description: Brief description shown in /help (required)
argument-hint: <placeholder> (optional, if command takes arguments)
---每个斜杠命令都是一个Markdown文件,结构如下:
markdown
---
description: 显示在/help中的简短描述(必填)
argument-hint: <占位符>(可选,如果命令接受参数)
---Command Title
命令标题
[Detailed instructions for the agent to execute autonomously]
undefined[供Agent自主执行的详细说明]
undefinedCommand Creation Workflow
命令创建工作流
Step 1: Determine Location
步骤1:确定存储位置
Auto-detect the appropriate location:
- Check git repository status:
git rev-parse --is-inside-work-tree 2>/dev/null - Default location:
- If in git repo → Project-level:
.claude/commands/ - If not in git repo → Global:
~/.claude/commands/
- If in git repo → Project-level:
- Allow user override:
- If user explicitly mentions "global" or "user-level" → Use
~/.claude/commands/ - If user explicitly mentions "project" or "project-level" → Use
.claude/commands/
- If user explicitly mentions "global" or "user-level" → Use
Report the chosen location to the user before proceeding.
自动检测合适的位置:
- 检查Git仓库状态:
git rev-parse --is-inside-work-tree 2>/dev/null - 默认位置:
- 如果在Git仓库内 → 项目级:
.claude/commands/ - 如果不在Git仓库内 → 全局:
~/.claude/commands/
- 如果在Git仓库内 → 项目级:
- 允许用户覆盖:
- 如果用户明确提到“global”或“user-level” → 使用
~/.claude/commands/ - 如果用户明确提到“project”或“project-level” → 使用
.claude/commands/
- 如果用户明确提到“global”或“user-level” → 使用
在继续之前,将选定的位置告知用户。
Step 2: Show Command Patterns
步骤2:展示命令模式
Help the user understand different command types. Load references/patterns.md to see available patterns:
- Workflow Automation - Analyze → Act → Report (e.g., submit-stack)
- Iterative Fixing - Run → Parse → Fix → Repeat (e.g., ensure-ci)
- Agent Delegation - Context → Delegate → Iterate (e.g., create-implementation-plan)
- Simple Execution - Run command with args (e.g., codex-review)
Ask the user: "Which pattern is closest to what you want to create?" This helps frame the conversation.
帮助用户了解不同的命令类型。加载references/patterns.md查看可用模式:
- 工作流自动化 - 分析 → 执行 → 报告(例如submit-stack)
- 迭代修复 - 运行 → 解析 → 修复 → 重复(例如ensure-ci)
- Agent委托 - 上下文 → 委托 → 迭代(例如create-implementation-plan)
- 简单执行 - 带参数运行命令(例如codex-review)
询问用户:“哪种模式最接近你想要创建的命令?”这有助于引导对话。
Step 3: Gather Command Information
步骤3:收集命令信息
Ask the user for key information:
向用户询问关键信息:
A. Command Name and Purpose
A. 命令名称与用途
Ask:
- "What should the command be called?" (for filename)
- "What does this command do?" (for description field)
Guidelines:
- Command names MUST be kebab-case (hyphens, NOT underscores)
- ✅ CORRECT: ,
submit-stack,ensure-cicreate-from-plan - ❌ WRONG: ,
submit_stack,ensure_cicreate_from_plan
- ✅ CORRECT:
- File names match command names: → invoked as
my-command.md/my-command - Description should be concise, action-oriented (appears in output)
/help
询问:
- “命令应该叫什么名字?”(用于文件名)
- “这个命令的作用是什么?”(用于description字段)
指南:
- 命令名称必须使用短横线分隔式(kebab-case,即使用连字符,而非下划线)
- ✅ 正确:,
submit-stack,ensure-cicreate-from-plan - ❌ 错误:,
submit_stack,ensure_cicreate_from_plan
- ✅ 正确:
- 文件名与命令名称一致:→ 调用方式为
my-command.md/my-command - 描述应简洁、以行动为导向(会显示在输出中)
/help
B. Arguments
B. 参数
Ask:
- "Does this command take any arguments?"
- "Are arguments required or optional?"
- "What should arguments represent?"
If command takes arguments:
- Add to frontmatter
argument-hint: <placeholder> - Use for required arguments
<angle-brackets> - Use for optional arguments
[square-brackets]
询问:
- “这个命令是否接受参数?”
- “参数是必填还是可选?”
- “参数代表什么内容?”
如果命令接受参数:
- 在前置元数据中添加
argument-hint: <占位符> - 使用表示必填参数
<尖括号> - 使用表示可选参数
[方括号]
C. Workflow Steps
C. 工作流步骤
Ask:
- "What are the specific steps this command should follow?"
- "What order should they happen in?"
- "What tools or commands should be used?"
Gather details about:
- Initial analysis or checks to perform
- Main actions to take
- How to handle results
- Success criteria
- Error handling approach
询问:
- “这个命令应该遵循哪些具体步骤?”
- “步骤的执行顺序是什么?”
- “应该使用哪些工具或命令?”
收集以下细节:
- 要执行的初始分析或检查
- 主要操作
- 如何处理结果
- 成功标准
- 错误处理方式
D. Tool Restrictions and Guidance
D. 工具限制与指导
Ask:
- "Should this command use any specific agents or tools?"
- "Are there any tools or operations it should avoid?"
- "Should it read any specific files for context?"
询问:
- “这个命令是否应该使用特定的Agent或工具?”
- “是否有任何工具或操作是需要避免的?”
- “它是否需要读取特定文件以获取上下文?”
Step 4: Generate Optimized Command
步骤4:生成优化的命令
Create the command file with agent-optimized instructions. Load references/best-practices.md for:
- Template structure
- Best practices for agent execution
- Writing style guidelines
- Quality checklist
Key principles:
- Use imperative/infinitive form (verb-first instructions)
- Be explicit and specific
- Include expected outcomes
- Provide concrete examples
- Define clear error handling
创建包含Agent优化执行说明的命令文件。加载references/best-practices.md获取以下内容:
- 模板结构
- Agent执行的最佳实践
- 写作风格指南
- 质量检查表
核心原则:
- 使用祈使/不定式形式(以动词开头的说明)
- 明确且具体
- 包含预期结果
- 提供具体示例
- 定义清晰的错误处理
Step 5: Create the Command File
步骤5:创建命令文件
-
Determine full file path:
- Project:
.claude/commands/[command-name].md - Global:
~/.claude/commands/[command-name].md
- Project:
-
Ensure directory exists:bash
mkdir -p [directory-path] -
Write the command file using the Write tool
-
Confirm with user:
- Report the file location
- Summarize what the command does
- Explain how to use it:
/command-name [arguments]
-
确定完整文件路径:
- 项目级:
.claude/commands/[command-name].md - 全局级:
~/.claude/commands/[command-name].md
- 项目级:
-
确保目录存在:bash
mkdir -p [directory-path] -
使用Write工具写入命令文件
-
与用户确认:
- 告知文件位置
- 总结命令的作用
- 说明使用方式:
/command-name [arguments]
Step 6: Test and Iterate (Optional)
步骤6:测试与迭代(可选)
If the user wants to test:
- Suggest testing:
You can test this command by running: /command-name [arguments] - Be ready to iterate based on feedback
- Update the file with improvements as needed
如果用户想要测试:
- 建议测试:
你可以通过运行以下命令测试此命令:/command-name [arguments] - 随时根据反馈进行迭代
- 根据需要更新文件以优化内容
Quick Tips
快速提示
For detailed guidance, load the bundled references:
- Load references/patterns.md when designing the command workflow
- Load references/examples.md to see how existing commands are structured
- Load references/best-practices.md before finalizing to ensure quality
Common patterns to remember:
- Use Bash tool for ,
pytest,pyright,ruff,prettier,makecommandsgt - Use Task tool to invoke subagents for specialized tasks
- Check for specific files first (e.g., ) before proceeding
.PLAN.md - Mark todos complete immediately, not in batches
- Include explicit error handling instructions
- Define clear success criteria
如需详细指导,请加载附带的参考文档:
- 设计命令工作流时,加载references/patterns.md
- 查看现有命令结构时,加载references/examples.md
- 最终确定前,加载references/best-practices.md以确保质量
需要记住的常见模式:
- 对、
pytest、pyright、ruff、prettier、make命令使用Bash工具gt - 使用Task工具调用子Agent以执行专业任务
- 先检查特定文件(例如),再继续执行后续步骤
.PLAN.md - 立即标记待办事项为完成,而非批量处理
- 包含明确的错误处理说明
- 定义清晰的成功标准
Summary
总结
When creating a command:
- Detect location (project vs global)
- Show patterns to frame the conversation
- Gather information (name, purpose, arguments, steps, tools)
- Generate optimized command with agent-executable instructions
- Create file at appropriate location
- Confirm and iterate as needed
Focus on creating commands that agents can execute autonomously, with clear steps, explicit tool usage, and proper error handling.
创建命令时:
- 检测位置(项目级 vs 全局级)
- 展示模式以引导对话
- 收集信息(名称、用途、参数、步骤、工具)
- 生成优化命令,包含Agent可自主执行的说明
- 在合适位置创建文件
- 确认并按需迭代
重点创建Agent可自主执行的命令,包含清晰的步骤、明确的工具使用说明和完善的错误处理机制。