Back to Details

pm-skill-creator

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Purpose

用途

Walk through the full skill design process interactively — from raw idea or content to a structured, repo-compliant SKILL.md draft. Asks adaptive questions to determine skill type, scope, structure, and content, then generates a ready-to-validate draft.
This skill is the conversational complement to 
skill-authoring-workflow
. That skill defines the process and validation gates. This one sits with you and figures out what to build before you build it.
通过交互式方式完成完整的Skill设计流程——从原始想法或内容到结构化、符合仓库规范的SKILL.md草稿。会提出适应性问题来确定Skill类型、范围、结构和内容，然后生成可直接验证的草稿。
本Skill是
skill-authoring-workflow
的对话式补充。那个Skill定义了流程和验证关卡，而本Skill会在你开始构建之前，与你一同确定要构建的内容。

Key Concepts

核心概念

When to Use This vs. Other Authoring Tools

何时使用本工具 vs 其他创作工具

ToolBest When
This skill (skill-creator)You have an idea or raw content and need help shaping it into the right structure through conversation
skill-authoring-workflow
You already know what to build and need the process checklist and validation steps
scripts/build-a-skill.sh
You know the structure and want a terminal wizard to collect sections
scripts/add-a-skill.sh
You have a source document and want AI-assisted generation end-to-end
工具适用场景
本Skill (skill-creator)你有想法或原始内容，需要通过对话帮助梳理成正确结构时
skill-authoring-workflow
你已经明确要构建的内容，需要流程清单和验证步骤时
scripts/build-a-skill.sh
你清楚结构，想要通过终端向导收集各部分内容时
scripts/add-a-skill.sh
你有源文档，想要端到端的AI辅助生成时

The Three Skill Types (Decision Criteria)

三种Skill类型（判定标准）

  • Component: One artifact or template. Self-contained. Gets referenced by other skills. Ask: "Is this a thing someone creates?"
  • Interactive: Guided conversation with adaptive questions and enumerated recommendations. Ask: "Does this require back-and-forth to be useful?"
  • Workflow: Multi-phase orchestration referencing other skills. Ask: "Does this coordinate multiple activities across steps?"
  • Component（组件型）：单个工件或模板，独立完整，可供其他Skill引用。自问：“这是用户会创建的某个具体事物吗？”
  • Interactive（交互式）：包含引导式对话、适应性问题和明确建议。自问：“它需要双向互动才能发挥作用吗？”
  • Workflow（工作流型）：多阶段编排，会引用其他Skill。自问：“它需要跨步骤协调多项活动吗？”

Skill Anatomy (Non-Negotiable Sections)

Skill的必备结构（不可缺少的章节）

Every skill requires these sections in order:
  1. Purpose — What it does + when to use it (outcome-focused)
  2. Key Concepts — Frameworks, definitions, mental models
  3. Application — Step-by-step instructions an agent can follow
  4. Examples — At least one concrete, specific example
  5. Common Pitfalls — Named failure modes with consequences and fixes
  6. References — Related skills, external frameworks, source material
每个Skill都必须包含以下章节，顺序固定：
  1. Purpose（用途） — 功能+适用场景（聚焦成果）
  2. Key Concepts（核心概念） — 框架、定义、思维模型
  3. Application（应用方法） — Agent可遵循的分步说明
  4. Examples（示例） — 至少一个具体、明确的示例
  5. Common Pitfalls（常见误区） — 命名明确的失败模式，包含后果和修复方案
  6. References（参考资料） — 相关Skill、外部框架、源材料

Metadata Constraints

元数据约束

  • name
    : lowercase kebab-case, ≤ 64 characters
  • description
    : ≤ 200 characters, must include trigger cue ("Use when...")
  • intent
    : longer repo-facing summary (no character limit)
  • type
    : one of 
    component
    , 
    interactive
    , 
    workflow
  • name
    ：小写短横线命名法（kebab-case），≤64字符
  • description
    ：≤200字符，必须包含触发提示（“Use when...”）
  • intent
    ：面向仓库的详细概述（无字符限制）
  • type
    ：必须是
    component
    、
    interactive
    、
    workflow
    三者之一

Facilitation Source of Truth

协作交互的基准规范

Use 
workshop-facilitation
 as the default interaction protocol for this skill.
It defines:
  • Session heads-up + entry mode (Guided, Context dump, Best guess)
  • One-question turns with plain-language prompts
  • Progress labels (e.g., Context Q1/5)
  • Interruption handling and pause/resume behavior
  • Numbered recommendations at decision points
  • Quick-select numbered response options (include 
    Other (specify)
    when useful)
This file defines the domain-specific content. If there is a conflict, follow this file's domain logic.
默认使用
workshop-facilitation
作为本Skill的交互协议。
它定义了：
  • 会话提示+进入模式（引导式、上下文导入、最佳猜测）
  • 单轮一问的简洁提示
  • 进度标签（如：Context Q1/5）
  • 中断处理和暂停/恢复机制
  • 决策点的编号建议
  • 快速选择的编号响应选项（必要时包含“其他（请说明）”）
本文档定义了领域特定内容。若存在冲突，以本文档的领域逻辑为准。

Application

应用方法

This interactive skill asks up to 5 adaptive questions, then delivers a complete SKILL.md draft with frontmatter, all required sections, and repo-compliant structure.
本交互式Skill最多提出5个适应性问题，然后交付完整的SKILL.md草稿，包含前置元数据、所有必备章节和符合仓库规范的结构。

Step 0: Session Start

步骤0：会话启动

Agent says:
"I'll help you design a new PM skill from scratch. This takes about 10-15 minutes and up to 5 questions. How do you want to start?
  1. Guided — I'll ask questions one at a time and build the skill from your answers (recommended)
  2. Context dump — Paste your raw content, notes, or framework and I'll propose a skill structure
  3. Best guess — Tell me just the topic and I'll draft a skill you can refine"
Agent话术：
“我将帮你从零开始设计新的PM Skill。整个过程约10-15分钟，最多包含5个问题。你想如何开始？
  1. 引导式 — 我会逐个提问，根据你的回答构建Skill（推荐）
  2. 上下文导入 — 粘贴你的原始内容、笔记或框架，我会提出Skill结构方案
  3. 最佳猜测 — 只需告诉我主题，我会生成可供你优化的Skill草稿”

Question 1: What's the Raw Material? (Q1/5)

问题1：原始素材是什么？（Q1/5）

Agent asks:
"What are we turning into a skill? Give me whatever you have."
  1. A framework or mental model — A structured way of thinking about a PM problem (e.g., prioritization matrix, decision tree)
  2. A template or artifact — A deliverable PMs create (e.g., PRD section, positioning statement, epic format)
  3. A process or workflow — A multi-step method for completing a PM task (e.g., discovery sprint, roadmap planning)
  4. A coaching or advisory topic — A domain where PMs need guided, adaptive help (e.g., stakeholder navigation, pricing decisions)
Or describe what you have in your own words.
Agent note: If the user pastes raw content instead of choosing an option, analyze the content and infer the answer. Confirm your interpretation before proceeding.
Agent提问：
“我们要把什么转化为Skill？请提供你已有的内容。”
  1. 框架或思维模型 — 解决PM问题的结构化思考方式（如：优先级矩阵、决策树）
  2. 模板或工件 — PM需要创建的交付物（如：PRD章节、定位声明、史诗格式）
  3. 流程或工作流 — 完成PM任务的多步骤方法（如：发现冲刺、路线图规划）
  4. 辅导或咨询主题 — PM需要引导式、适应性帮助的领域（如：利益相关方沟通、定价决策）
或者用你自己的语言描述你拥有的内容。
Agent注意事项： 如果用户粘贴原始内容而非选择选项，需分析内容并推断答案，在继续前确认你的解读。

Question 2: Skill Type Decision (Q2/5)

问题2：Skill类型判定（Q2/5）

Based on Q1 answer, agent recommends a type and confirms:
If Q1 = Framework or Template:
"This sounds like a component skill — a self-contained artifact or reference. It would include a template, quality criteria, and examples. Does that fit, or is there a conversational/adaptive element I'm missing?
  1. Yes, component — It's a standalone deliverable or reference
  2. Actually, it needs conversation — Users need guided questions to use it well (→ interactive)
  3. It's bigger than one artifact — It orchestrates multiple steps or other skills (→ workflow)"
If Q1 = Process or Workflow:
"This sounds like a workflow skill — a multi-phase process. Does it reference or orchestrate other discrete skills/artifacts, or is it more of a guided conversation?
  1. Yes, workflow — It has distinct phases and may reference other skills
  2. It's more conversational — The value comes from adaptive questions and recommendations (→ interactive)
  3. It's simpler than I described — Really it's one artifact with steps (→ component)"
If Q1 = Coaching or Advisory:
"This sounds like an interactive skill — guided questions that adapt based on answers, ending with enumerated recommendations. Sound right?
  1. Yes, interactive — It needs back-and-forth to be useful
  2. It's more of a reference — Users just need the framework, not a conversation (→ component)
  3. It's a full process — Multiple phases, orchestrates other skills (→ workflow)"
基于问题1的答案，Agent推荐类型并确认：
若问题1=框架或模板：
“这听起来是一个component型Skill——独立完整的工件或参考资料。它会包含模板、质量标准和示例。是否符合你的需求，还是我忽略了需要对话/互动的部分？
  1. 是的，component型 — 它是独立的交付物或参考资料
  2. 实际上需要对话 — 用户需要引导式提问才能更好地使用它（→ interactive型）
  3. 它比单个工件更复杂 — 它需要协调多个步骤或其他Skill（→ workflow型）”
若问题1=流程或工作流：
“这听起来是一个workflow型Skill——多阶段流程。它是否引用或编排其他独立的Skill/工件，还是更偏向引导式对话？
  1. 是的，workflow型 — 它有明确的阶段，可能引用其他Skill
  2. 更偏向对话式 — 核心价值来自适应性问题和建议（→ interactive型）
  3. 比我描述的更简单 — 实际上是带步骤的单个工件（→ component型）”
若问题1=辅导或咨询主题：
“这听起来是一个interactive型Skill——根据回答调整的引导式提问，最终给出编号建议。对吗？
  1. 是的，interactive型 — 需要双向互动才能发挥作用
  2. 更偏向参考资料 — 用户只需要框架，不需要对话（→ component型）
  3. 是完整的流程 — 多阶段，编排其他Skill（→ workflow型）”

Question 3: Scope and Naming (Q3/5)

问题3：范围与命名（Q3/5）

Agent asks:
"What should we call this skill? I need two things:
a) A working name — lowercase-kebab-case, max 64 characters (e.g., 
feature-investment-advisor
, 
user-story
, 
discovery-process
)
b) A one-sentence description — What it does + when to use it. Must fit in 200 characters and start with a verb or 'Use when...'
Give me your best attempt and I'll tighten it if needed. Or just describe the skill's purpose and I'll propose both."
Agent note: Validate the name format (kebab-case, ≤ 64 chars) and description length (≤ 200 chars) before proceeding. If either fails, suggest a fix.
Agent提问：
“这个Skill应该叫什么？我需要两个信息：
a) 工作名称 — 小写短横线命名法（kebab-case），最多64字符（如：
feature-investment-advisor
, 
user-story
, 
discovery-process
）
b) 一句话描述 — 功能+适用场景。必须≤200字符，以动词或‘Use when...’开头
请给出你的初步想法，我会帮你优化。或者只需描述Skill的用途，我来提出两者的方案。”
Agent注意事项： 在继续前验证名称格式（kebab-case、≤64字符）和描述长度（≤200字符）。若不符合，建议修改方案。

Question 4: Key Content (Q4/5)

问题4：核心内容（Q4/5）

This question adapts based on skill type from Q2:
For component skills:
"What are the core elements of this artifact or framework? I need:
  1. The template or structure — What sections/fields does it contain?
  2. Quality criteria — What separates a good one from a bad one?
  3. One concrete example — A filled-in version showing it done well
Give me whatever you have — bullet points, rough notes, or a full draft."
For interactive skills:
"Walk me through the conversation flow:
  1. What's the opening question? — What does the user need to tell you first?
  2. What are the 2-4 branching paths? — How do answers change what comes next?
  3. What recommendations emerge? — What are the 3-5 outcomes you'd offer?
Give me the decision tree as you see it — even if it's rough."
For workflow skills:
"Map out the phases:
  1. What are the major steps? (Usually 3-6 phases)
  2. What's the input and output of each phase?
  3. Which existing skills does this reference? (Check with 
    scripts/find-a-skill.sh --keyword <topic>
    )
  4. Where are the decision points?
Give me the flow — sequential, branching, or both."
本问题会根据问题2确定的Skill类型调整：
针对component型Skill：
“这个工件或框架的核心元素是什么？我需要：
  1. 模板或结构 — 包含哪些章节/字段？
  2. 质量标准 — 优秀版本与糟糕版本的区别是什么？
  3. 一个具体示例 — 填写完整的优秀示例
请提供你已有的内容—— bullet点、粗略笔记或完整草稿均可。”
针对interactive型Skill：
“请梳理对话流程：
  1. 开场问题是什么？ — 用户首先需要告诉你什么？
  2. 2-4个分支路径是什么？ — 回答如何影响后续内容？
  3. 会生成哪些建议？ — 你会提供3-5种什么结果？
请给出你设想的决策树——即使是粗略的版本也可以。”
针对workflow型Skill：
“梳理各阶段：
  1. 主要步骤有哪些？（通常3-6个阶段）
  2. 每个阶段的输入和输出是什么？
  3. 会引用哪些现有Skill？（可使用
    scripts/find-a-skill.sh --keyword <topic>
    查询）
  4. 决策点在哪里？
请给出流程——顺序式、分支式或两者结合均可。”

Question 5: Pitfalls and Edge Cases (Q5/5)

问题5：误区与边缘情况（Q5/5）

Agent asks:
"What goes wrong when people do this badly? I need 2-3 failure modes:
  1. Name the failure — Give it a label (e.g., 'Metrics Theater', 'Hero Syndrome')
  2. Describe the consequence — What happens when someone falls into this trap?
  3. State the fix — What's the corrective action?
If you're not sure, tell me the most common mistake you've seen and I'll help structure it."
Agent提问：
“人们操作不当会出现什么问题？我需要2-3种失败模式：
  1. 命名失败模式 — 给它一个标签（如：‘Metrics Theater’、‘Hero Syndrome’）
  2. 描述后果 — 陷入这个陷阱会发生什么？
  3. 说明修复方案 — 纠正措施是什么？
如果你不确定，请告诉我你见过的最常见错误，我会帮你梳理结构。”

Draft Generation

草稿生成

After collecting answers to Q1-Q5, the agent generates a complete SKILL.md draft including:
  1. YAML frontmatter — 
    name
    , 
    description
    , 
    intent
    , 
    type
    , 
    best_for
    , 
    scenarios
    , 
    estimated_time
  2. Purpose — Synthesized from Q1 + Q3
  3. Key Concepts — Structured from Q4 content
  4. Application — Step-by-step instructions derived from Q4
  5. Examples — Concrete example from Q4 (or generated if not provided)
  6. Common Pitfalls — Structured from Q5
  7. References — Related skills identified during conversation + source material
Agent says after generating:
"Here's your draft SKILL.md. Before we call it done:
  1. Review the draft — Does it capture your intent?
  2. Run validation — 
    python3 scripts/check-skill-metadata.py skills/<name>/SKILL.md
  3. Check triggers — 
    python3 scripts/check-skill-triggers.py skills/<name>/SKILL.md --show-cases
  4. Smoke test — 
    ./scripts/test-a-skill.sh --skill <name> --smoke
Want me to adjust anything, or are you ready to validate?"
收集问题1-5的答案后，Agent会生成完整的SKILL.md草稿，包含：
  1. YAML前置元数据 — 
    name
    , 
    description
    , 
    intent
    , 
    type
    , 
    best_for
    , 
    scenarios
    , 
    estimated_time
  2. Purpose（用途） — 从问题1+问题3整合而来
  3. Key Concepts（核心概念） — 从问题4的内容结构化而来
  4. Application（应用方法） — 从问题4推导的分步说明
  5. Examples（示例） — 来自问题4的具体示例（若未提供则自动生成）
  6. Common Pitfalls（常见误区） — 从问题5结构化而来
  7. References（参考资料） — 对话中识别的相关Skill+源材料
Agent生成草稿后的话术：
“这是你的SKILL.md草稿。在完成之前：
  1. 审核草稿 — 是否符合你的意图？
  2. 运行验证 — 
    python3 scripts/check-skill-metadata.py skills/<name>/SKILL.md
  3. 检查触发条件 — 
    python3 scripts/check-skill-triggers.py skills/<name>/SKILL.md --show-cases
  4. 冒烟测试 — 
    ./scripts/test-a-skill.sh --skill <name> --smoke
需要我调整内容，还是你准备好验证了？”

Examples

示例

Example: Framework → Component Skill

示例：框架 → Component型Skill

Q1: "1 — A framework. It's a 2x2 matrix for evaluating build-vs-buy decisions." Q2: "1 — Yes, component." Q3: "Name: 
build-vs-buy-matrix
. Description: 'Evaluate build-vs-buy decisions using a 2x2 matrix of strategic value and implementation complexity. Use when choosing between building internally and buying a solution.'" Q4: "The axes are strategic differentiation (low/high) and implementation complexity (low/high). Four quadrants: Buy (low/low), Partner (low/high), Build (high/low), Invest (high/high). Quality criteria: each quadrant has a clear decision rule and one example." Q5: "Biggest mistake: treating it as a one-time decision. Consequence: you build something, market shifts, and you're stuck maintaining undifferentiated tech. Fix: revisit quarterly."
Output: Complete component skill with template, filled example, and the quarterly-review pitfall.
问题1： “1 — 一个框架。用于评估自研vs采购决策的2x2矩阵。” 问题2： “1 — 是的，component型。” 问题3： “名称：
build-vs-buy-matrix
。描述：‘使用战略价值和实施复杂度的2x2矩阵评估自研vs采购决策。当选择内部自研还是采购解决方案时使用。’” 问题4： “坐标轴是战略差异化（低/高）和实施复杂度（低/高）。四个象限：采购（低/低）、合作（低/高）、自研（高/低）、投入（高/高）。质量标准：每个象限有明确的决策规则和一个示例。” 问题5： “最大的错误：将其视为一次性决策。后果：你自研了某个产品，市场变化后，不得不维护无差异化的技术。修复方案：每季度重新评估。”
输出： 完整的component型Skill，包含模板、填写好的示例和“季度重评估”误区。

Example: Advisory Topic → Interactive Skill

示例：咨询主题 → Interactive型Skill

Q1: "4 — Coaching topic. Helping PMs navigate their first pricing conversation with leadership." Q2: "1 — Yes, interactive." Q3: "Name: 
pricing-conversation-advisor
. Description: 'Prepare for your first pricing conversation with leadership. Use when presenting pricing changes or new monetization to executives.'" Q4: "Opening question: What's the pricing change? Branches: new product pricing, price increase on existing, freemium-to-paid conversion, usage-based transition. Recommendations: data to bring, objection handling, framing strategies." Q5: "Failure mode: 'Leading with the Number.' Consequence: executives fixate on the price point instead of the strategy. Fix: always lead with the business case and customer impact before revealing the number."
Output: Complete interactive skill with 3-question flow, 4 branches, and enumerated recommendations per branch.
问题1： “4 — 辅导主题。帮助PM首次与领导层进行定价对话。” 问题2： “1 — 是的，interactive型。” 问题3： “名称：
pricing-conversation-advisor
。描述：‘为你首次与领导层的定价对话做准备。当向高管展示定价变更或新 monetization方案时使用。’” 问题4： “开场问题：定价变更内容是什么？分支：新产品定价、现有产品提价、免费转付费、基于使用量的转型。建议：需准备的数据、异议处理、框架策略。” 问题5： “失败模式：‘先报数字’。后果：高管只关注价格点而非策略。修复方案：在公布数字前，始终先展示业务案例和客户影响。”
输出： 完整的interactive型Skill，包含3个问题的流程、4个分支，以及每个分支的编号建议。

Anti-Pattern: Scope Creep

反模式：范围蔓延

Q1: "I want a skill that covers all of product strategy — vision, roadmaps, OKRs, prioritization, stakeholder alignment, and metrics."
Agent response: "That's 4-6 separate skills, not one. Let's pick the one that matters most right now and build that. Which piece is most urgent?
  1. Vision → component skill
  2. Roadmap planning → workflow skill (already exists: 
    roadmap-planning
    )
  3. Prioritization → interactive skill (already exists: 
    prioritization-advisor
    )
  4. Something else
Start with one. We can build the others after."
问题1： “我想要一个涵盖所有产品策略的Skill——愿景、路线图、OKR、优先级、利益相关方对齐和指标。”
Agent回复： “这是4-6个独立的Skill，不是一个。我们先选当前最关键的部分来构建。哪部分最紧急？
  1. 愿景 → component型Skill
  2. 路线图规划 → workflow型Skill（已存在：
    roadmap-planning
    ）
  3. 优先级 → interactive型Skill（已存在：
    prioritization-advisor
    ）
  4. 其他内容
先从一个开始，之后可以构建其他的。”

Common Pitfalls

常见误区

Pitfall 1: Type Confusion

误区1：类型混淆

Symptom: Calling something a "workflow" when it's really a template with numbered steps.
Consequence: Over-engineered skill that references non-existent sub-skills and confuses users.
Fix: Apply the type decision questions from Q2 honestly. If it doesn't orchestrate other skills or have distinct phases with different inputs/outputs, it's probably a component.
症状： 将带编号步骤的模板称为“workflow”。
后果： Skill设计过度，引用不存在的子Skill，使用户困惑。
修复方案： 诚实地应用问题2中的类型判定问题。如果它不需要编排其他Skill，也没有不同输入/输出的明确阶段，那它很可能是component型。

Pitfall 2: Description That Doesn't Trigger

误区2：描述无触发点

Symptom: Description says what the skill is ("A framework for pricing decisions") but not when to use it.
Consequence: Claude can't match user requests to the skill. It sits unused.
Fix: Every description must answer "Use when..." — e.g., "Evaluate pricing decisions using a structured framework. Use when choosing between pricing models or preparing a pricing proposal."
症状： 描述只说明Skill是什么（“定价决策框架”），但未说明何时使用。
后果： Claude无法将用户请求与该Skill匹配，导致Skill被闲置。
修复方案： 每个描述必须回答“Use when...”——例如：“使用结构化框架评估定价决策。当选择定价模型或准备定价提案时使用。”

Pitfall 3: Skipping Validation

误区3：跳过验证

Symptom: "The draft looks good, let's ship it."
Consequence: Broken frontmatter, missing sections, failed cross-references, inconsistent catalog counts.
Fix: Always run 
check-skill-metadata.py
and 
check-skill-triggers.py
before considering the skill done. No exceptions.
症状： “草稿看起来不错，我们发布吧。”
后果： 前置元数据错误、章节缺失、交叉引用失效、目录计数不一致。
修复方案： 在认为Skill完成前，必须运行
check-skill-metadata.py
和
check-skill-triggers.py
，无例外。

Pitfall 4: Kitchen Sink Scope

误区4：范围过大

Symptom: Trying to pack an entire domain into one skill.
Consequence: A 500-line monster that does nothing well. Too broad to trigger accurately, too long to be useful.
Fix: One skill = one job. If you need more than 6 application steps or more than 4 branches, you probably need multiple skills.
症状： 试图将整个领域塞进一个Skill。
后果： 生成500行的臃肿Skill，无法很好地完成任何功能。范围太广导致触发不准确，篇幅太长难以使用。
修复方案： 一个Skill只做一件事。如果需要超过6个应用步骤或4个分支，你可能需要多个Skill。

References

参考资料

Related Skills

相关Skill

  • skill-authoring-workflow
     — The process checklist and validation gates; use after this skill generates a draft
  • workshop-facilitation
     — Facilitation protocol for this interactive skill
  • skill-authoring-workflow
     — 流程清单和验证关卡；在本Skill生成草稿后使用
  • workshop-facilitation
     — 本交互式Skill的协作交互协议

Repo Tools

仓库工具

  • scripts/build-a-skill.sh
    — Terminal wizard for section-by-section skill creation
  • scripts/add-a-skill.sh
    — Content-first automated skill generator
  • scripts/check-skill-metadata.py
    — Structural validation
  • scripts/check-skill-triggers.py
    — Trigger-readiness audit
  • scripts/test-a-skill.sh
    — Full quality gate
  • scripts/find-a-skill.sh
    — Check for overlapping skills before creating
  • scripts/build-a-skill.sh
    — 用于分章节创建Skill的终端向导
  • scripts/add-a-skill.sh
    — 基于内容的自动化Skill生成工具
  • scripts/check-skill-metadata.py
    — 结构验证工具
  • scripts/check-skill-triggers.py
    — 触发就绪性审核工具
  • scripts/test-a-skill.sh
    — 完整质量关卡工具
  • scripts/find-a-skill.sh
    — 创建前检查是否有重叠Skill

Documentation

文档

  • CLAUDE.md
    — Master skill distillation protocol and quality standards
  • docs/Building PM Skills.md
    — Manual skill creation guide
  • docs/Add-a-Skill Utility Guide.md
    — Automated creation guide
  • CLAUDE.md
    — 核心Skill提炼协议和质量标准
  • docs/Building PM Skills.md
    — 手动创建Skill指南
  • docs/Add-a-Skill Utility Guide.md
    — 自动化创建指南