Back to Details

skill-developer

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Skill Developer

技能开发者

Meta-skill for creating new Claude Code skills, including skills that wrap MCP pipelines.
用于创建新Claude Code技能的元技能，包括封装MCP管道的技能。

When to Use

使用场景

  • "Create a skill for X"
  • "Help me make a new skill"
  • "Turn this script into a skill"
  • "How do I create a skill?"
  • "为X创建一个技能"
  • "帮我制作一个新技能"
  • "把这个脚本转换成技能"
  • "我该如何创建技能？"

Skill Structure

技能结构

Skills live in 
.claude/skills/<skill-name>/
:
.claude/skills/my-skill/
├── SKILL.md          # Required: Main skill definition
├── scripts/          # Optional: Supporting scripts
└── templates/        # Optional: Templates, examples
技能存储在
.claude/skills/<skill-name>/
目录下：
.claude/skills/my-skill/
├── SKILL.md          # 必填：主要技能定义
├── scripts/          # 可选：支持脚本
└── templates/        # 可选：模板、示例

SKILL.md Format

SKILL.md格式

yaml
---
name: skill-name
description: Brief description (shown in skill list)
allowed-tools: [Bash, Read, Write]  # Optional: restrict tools
---
yaml
---
name: skill-name
description: 简短描述（显示在技能列表中）
allowed-tools: [Bash, Read, Write]  # 可选：限制工具
---

Skill Name

技能名称

When to Use

使用场景

[When Claude should discover this skill]
[Claude应触发该技能的场景]

Instructions

操作步骤

[Step-by-step instructions for Claude to follow]
[Claude需遵循的分步操作说明]

Examples

示例

[Usage examples]
undefined
[使用示例]
undefined

Creating an MCP Pipeline Skill

创建MCP管道技能

To create a new MCP chain script and wrap it as a skill:
要创建新的MCP链式脚本并将其封装为技能：

Step 1: Use the Template

步骤1：使用模板

Copy the multi-tool-pipeline template:
bash
cp $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py $CLAUDE_PROJECT_DIR/scripts/my_pipeline.py
Reference the template pattern:
bash
cat $CLAUDE_PROJECT_DIR/.claude/skills/multi-tool-pipeline/SKILL.md
cat $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py
复制多工具管道模板：
bash
cp $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py $CLAUDE_PROJECT_DIR/scripts/my_pipeline.py
参考模板示例：
bash
cat $CLAUDE_PROJECT_DIR/.claude/skills/multi-tool-pipeline/SKILL.md
cat $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py

Step 2: Customize the Script

步骤2：自定义脚本

Edit your new script to chain the MCP tools you need:
python
async def main():
    from runtime.mcp_client import call_mcp_tool
    args = parse_args()

    # Chain your MCP tools (serverName__toolName)
    result1 = await call_mcp_tool("server1__tool1", {"param": args.arg1})
    result2 = await call_mcp_tool("server2__tool2", {"input": result1})

    print(result2)
编辑新脚本以串联所需的MCP工具：
python
async def main():
    from runtime.mcp_client import call_mcp_tool
    args = parse_args()

    # 串联你的MCP工具（格式为serverName__toolName）
    result1 = await call_mcp_tool("server1__tool1", {"param": args.arg1})
    result2 = await call_mcp_tool("server2__tool2", {"input": result1})

    print(result2)

Step 2: Create the Skill

步骤2：创建技能

Create 
.claude/skills/my-pipeline/SKILL.md
:
markdown
---
name: my-pipeline
description: What the pipeline does
allowed-tools: [Bash, Read]
---
创建
.claude/skills/my-pipeline/SKILL.md
文件：
markdown
---
name: my-pipeline
description: 该管道的功能
allowed-tools: [Bash, Read]
---

My Pipeline Skill

我的管道技能

When to Use

使用场景

  • [Trigger conditions]
  • [触发条件]

Instructions

操作步骤

Run the pipeline:
```bash uv run python -m runtime.harness scripts/my_pipeline.py --arg1 "value" ```
运行管道：
```bash uv run python -m runtime.harness scripts/my_pipeline.py --arg1 "value" ```

Parameters

参数

  • --arg1
    : Description
  • --arg1
    : 参数说明

MCP Servers Required

所需MCP服务器

  • server1: For tool1
  • server2: For tool2
undefined
  • server1: 用于tool1
  • server2: 用于tool2
undefined

Step 3: Add Triggers (Optional)

步骤3：添加触发条件（可选）

Add to 
.claude/skills/skill-rules.json
:
json
{
  "skills": {
    "my-pipeline": {
      "type": "domain",
      "enforcement": "suggest",
      "priority": "medium",
      "description": "What it does",
      "promptTriggers": {
        "keywords": ["keyword1", "keyword2"],
        "intentPatterns": ["(pattern).*?(match)"]
      }
    }
  }
}
添加到
.claude/skills/skill-rules.json
文件中：
json
{
  "skills": {
    "my-pipeline": {
      "type": "domain",
      "enforcement": "suggest",
      "priority": "medium",
      "description": "功能说明",
      "promptTriggers": {
        "keywords": ["keyword1", "keyword2"],
        "intentPatterns": ["(pattern).*?(match)"]
      }
    }
  }
}

Reference Files

参考文件

For full details, read:
bash
cat $CLAUDE_PROJECT_DIR/.claude/rules/skill-development.md
cat $CLAUDE_PROJECT_DIR/.claude/rules/mcp-scripts.md
如需详细信息，请查看：
bash
cat $CLAUDE_PROJECT_DIR/.claude/rules/skill-development.md
cat $CLAUDE_PROJECT_DIR/.claude/rules/mcp-scripts.md

Quick Checklist

快速检查清单

  • SKILL.md has frontmatter (name, description)
  • "When to Use" section is clear
  • Instructions are copy-paste ready
  • MCP servers documented if needed
  • Triggers added to skill-rules.json (optional)
  • SKILL.md包含前置元数据（名称、描述）
  • "使用场景"部分清晰明确
  • 操作步骤可直接复制使用
  • 按需记录MCP服务器信息
  • 已将触发条件添加到skill-rules.json（可选）

Examples in This Repo

仓库中的示例

Look at existing skills for patterns:
bash
ls $CLAUDE_PROJECT_DIR/.claude/skills/
cat $CLAUDE_PROJECT_DIR/.claude/skills/commit/SKILL.md
cat $CLAUDE_PROJECT_DIR/.claude/skills/firecrawl-scrape/SKILL.md
查看现有技能以参考模式：
bash
ls $CLAUDE_PROJECT_DIR/.claude/skills/
cat $CLAUDE_PROJECT_DIR/.claude/skills/commit/SKILL.md
cat $CLAUDE_PROJECT_DIR/.claude/skills/firecrawl-scrape/SKILL.md