utility-pm-skill-builder
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chinese<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 -->
<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 -->
PM Skill Builder
PM 技能构建器
This skill creates new PM skills for the pm-skills library. It produces a
Skill Implementation Packet . a complete design document with draft files .
in a staging area for review before promotion to canonical locations.
本技能为pm-skills库创建新的PM技能。它会生成一个技能实现包——包含草稿文件的完整设计文档——并存储在暂存区,待审核通过后再推广到正式位置。
When to Use
适用场景
- When you have an idea for a new PM skill
- When you want to add a domain skill (phase-specific), foundation skill (cross-cutting), or utility skill (meta/tooling) to the pm-skills library
- When a contributor needs guided skill creation that follows repo conventions
- 当你有一个新的PM技能想法时
- 当你想要向pm-skills库中添加领域技能(特定阶段)、基础技能(跨阶段)或实用技能(元/工具类)时
- 当贡献者需要遵循仓库规范的引导式技能创建流程时
When NOT to Use
不适用场景
- To modify an existing skill → use a future validation/iteration utility (planned)
- To create a skill for a non-pm-skills context → use a general agent skill builder (planned)
- To create a workflow → workflows are authored directly, not via this builder
- 修改现有技能 → 使用未来的验证/迭代工具(已规划)
- 为非pm-skills场景创建技能 → 使用通用Agent技能构建器(已规划)
- 创建工作流 → 工作流需直接编写,不通过本构建器
Instructions
操作步骤
When asked to create a new PM skill, follow these steps:
当被要求创建新PM技能时,请遵循以下步骤:
Step 1: Understand the Idea
步骤1:理解需求
Accept the idea in either form:
- Problem-first: "What PM problem does this skill solve? Who runs into this problem, and what do they currently produce (or fail to produce)?"
- Skill-first: "Describe the skill you want to create. What artifact does it produce? What PM activity does it support?"
Both entry points produce the same downstream flow. If the user provides
one form, do not ask for the other . extract what you need and proceed.
If the idea is vague, ask ONE follow-up question to clarify the artifact
type and target audience before proceeding.
接受以下两种形式的想法:
- 问题导向:“该技能解决PM的什么问题?谁会遇到这个问题,他们当前产出了什么(或未能产出什么)?”
- 技能导向:“描述你想要创建的技能。它会产出什么成果?支持PM的哪些活动?”
两种入口都会进入相同的后续流程。如果用户提供了其中一种形式,无需询问另一种——提取所需信息后继续即可。
如果想法模糊,仅需提出一个跟进问题,明确成果类型和目标受众后再继续。
Step 2: Gap Analysis
步骤2:差距分析
Check ALL existing skills for overlap. Use the Current Library Reference
table below AND scan the directory for the latest inventory.
skills/Present findings with specificity:
- Name each overlapping skill and explain what it covers
- Identify the specific gap this new skill would fill
- If overlap is high, trigger the Why Gate (see below)
Why Gate (triggers when overlap is found):
Ask the user: "Name 2-3 specific prompts or scenarios where the existing
skills fail to produce what you need."
Kill Gate: If the user cannot articulate convincing gaps, recommend
an alternative:
- "Revise [existing skill] to cover this case"
- "Create a workflow combining [skill A] + [skill B]"
- "Add a command variant, not a new skill"
- "This is a documentation improvement, not a new skill"
Do not proceed past the kill gate without either convincing evidence of
a gap or explicit user override.
检查所有现有技能是否存在重叠。使用下方的当前库参考表并扫描目录获取最新技能清单。
skills/具体呈现分析结果:
- 列出每个重叠技能的名称并说明其覆盖范围
- 明确新技能将填补的具体差距
- 如果重叠度高,触发Why Gate(见下文)
Why Gate(发现重叠时触发):
询问用户:“请列举2-3个现有技能无法满足你需求的具体提示或场景。”
终止门:如果用户无法清晰阐述合理的差距,推荐替代方案:
- “修改[现有技能]以覆盖此场景”
- “创建结合[技能A] + [技能B]的工作流”
- “添加命令变体,而非新技能”
- “这属于文档改进,而非新技能”
若没有令人信服的差距证据或用户明确要求覆盖,请勿越过终止门。
Step 3: Scope Check
步骤3:范围检查
Evaluate whether the idea should be ONE skill or MULTIPLE skills.
Splitting signals:
- The idea produces multiple distinct artifact types
- The idea crosses Triple Diamond phases (e.g., Discover + Deliver)
- The description naturally contains "and" connecting two activities
If splitting is warranted, present the recommendation:
"This seems to cover two distinct PM activities:
- [Activity A] → produces [Artifact A]
- [Activity B] → produces [Artifact B] These work better as separate skills that can be chained via a workflow. Want to proceed with just [Activity A] for now?"
评估该想法应作为单个技能还是多个技能。
拆分信号:
- 该想法会产出多种不同类型的成果
- 该想法跨越了三重钻石阶段(例如:探索+交付)
- 描述中自然包含连接两个活动的“和”
如果需要拆分,提出建议:
“这似乎涵盖了两个不同的PM活动:
- [活动A] → 产出[成果A]
- [活动B] → 产出[成果B] 将它们拆分为可通过工作流串联的独立技能会更合适。 是否先从[活动A]开始推进?”
Step 4: Classification + Repo-Fit
步骤4:分类与仓库适配
Determine the skill's classification and naming:
Domain skills (phase-specific PM activities):
- Phase: discover | define | develop | deliver | measure | iterate
- Directory:
{phase}-{skill-name} - Frontmatter: (required), no
phase: {phase}fieldclassification
Foundation skills (cross-cutting, used across phases):
- No phase
- Directory:
foundation-{skill-name} - Frontmatter: (required), no
classification: foundationfieldphase - Use when: the skill applies to multiple phases equally
Utility skills (meta-skills, repo tooling):
- No phase
- Directory:
utility-{skill-name} - Frontmatter: (required), no
classification: utilityfieldphase - Use when: the skill operates on the repo, workflow, or other skills
Exemplar selection:
Identify 1-2 existing skills that are the closest structural match:
- Same phase > same category > similar artifact type
- Read their SKILL.md to understand section structure, instruction style, output contract format, and quality checklist pattern
- Name the exemplars explicitly: "Modeled after [skill] . same phase, [category] category"
Present the classification and exemplar selection for user confirmation.
确定技能的分类和命名:
领域技能(特定阶段的PM活动):
- 阶段:discover | define | develop | deliver | measure | iterate
- 目录:
{phase}-{skill-name} - 前置元数据:(必填),无
phase: {phase}字段classification
基础技能(跨阶段,适用于多个阶段):
- 无阶段
- 目录:
foundation-{skill-name} - 前置元数据:(必填),无
classification: foundation字段phase - 适用场景:该技能同等适用于多个阶段
实用技能(元技能、仓库工具):
- 无阶段
- 目录:
utility-{skill-name} - 前置元数据:(必填),无
classification: utility字段phase - 适用场景:该技能作用于仓库、工作流或其他技能
范例选择:
找出1-2个结构最匹配的现有技能:
- 优先选择同阶段 > 同类别 > 相似成果类型
- 阅读它们的SKILL.md,了解章节结构、指令风格、输出契约格式和质量检查表模式
- 明确列出范例:“以[技能]为模板——同阶段,[类别]类”
向用户呈现分类结果和范例选择,待确认后继续。
Step 5: Generate Skill Implementation Packet
步骤5:生成技能实现包
Produce the complete packet using as the format.
The packet includes:
references/TEMPLATE.md- Decision . recommendation + Why Gate evidence (if applicable)
- Classification . type, phase (if domain), category, directory name
- Overlap Analysis . what was found, why this skill is still needed
- Quality Forecast . K/P/C/W zone distribution + writing guidance:
- Knowledge-heavy (≥35% K): reference frameworks, include When to Use
- Process-heavy (≥35% P): numbered steps, prescriptive, clear I/O per step
- Constraint-heavy (≥35% C): MUST/SHOULD/MUST NOT rules, separate section
- Wisdom-heavy (≥25% W): reflective questions, guide thinking
- Exemplar Skills . which existing skills modeled, why
- Draft Frontmatter . complete, valid YAML block
- Draft SKILL.md . full content (not an outline), mirroring exemplar structure
- Draft TEMPLATE.md . section headers with guidance comments
- Draft EXAMPLE.md . complete, realistic example (150-300 lines) with a specific PM scenario, every section filled, optional sections demonstrated both filled and skipped
- Draft Command . command frontmatter
- AGENTS.md Entry . exact text to add
- Validation Checklist . all CI rules checked against the draft
- Next Steps . local CI, testing, contribution workflow
使用作为格式,生成完整的实现包。包中包含:
references/TEMPLATE.md- 决策——建议+Why Gate证据(如适用)
- 分类——类型、阶段(若为领域技能)、类别、目录名称
- 重叠分析——发现的重叠内容、新技能仍需存在的原因
- 质量预测——K/P/C/W区域分布+写作指南:
- 知识密集型(≥35% K):参考框架,包含适用场景
- 流程密集型(≥35% P):编号步骤,规范性强,明确每步的输入/输出
- 约束密集型(≥35% C):MUST/SHOULD/MUST NOT规则,单独章节
- 智慧密集型(≥25% W):反思性问题,引导思考
- 范例技能——参考了哪些现有技能,原因是什么
- 草稿前置元数据——完整、有效的YAML块
- 草稿SKILL.md——完整内容(非大纲),镜像范例结构
- 草稿TEMPLATE.md——带指导注释的章节标题
- 草稿EXAMPLE.md——完整、真实的示例(150-300行),包含特定PM场景,填充所有章节,同时展示可选章节的填写和跳过情况
- 草稿命令——命令前置元数据
- AGENTS.md条目——需添加的准确文本
- 验证检查表——对照草稿检查所有CI规则
- 下一步——本地CI、测试、贡献流程
Step 6: Write to Staging Area
步骤6:写入暂存区
Write all generated files to the staging area:
_staging/pm-skill-builder/{skill-name}/
├── SKILL.md ← draft skill file
├── references/
│ ├── TEMPLATE.md ← draft template
│ └── EXAMPLE.md ← draft example
└── command.md ← draft commandNote:is gitignored . draft artifacts never ship in releases. The staging folder is discarded after promotion._staging/
Report what was written and where.
将所有生成的文件写入暂存区:
_staging/pm-skill-builder/{skill-name}/
├── SKILL.md ← 技能草稿文件
├── references/
│ ├── TEMPLATE.md ← 模板草稿
│ └── EXAMPLE.md ← 示例草稿
└── command.md ← 命令草稿注意:已被git忽略——草稿文件不会随版本发布。 推广后将删除暂存文件夹。_staging/
报告写入的内容和位置。
Step 7: Promote (on confirmation)
步骤7:推广(确认后)
Ask: "Review the packet above. When ready, I'll promote the files to
their canonical locations. Proceed? [yes/no]"
If yes, promote by copying each file from staging to its canonical path:
| Staging file | Canonical location |
|---|---|
| |
| |
| |
| |
Where is the classification-prefixed directory (e.g., ).
{dir-name}deliver-change-communicationThen:
- Create the target directory:
skills/{dir-name}/references/ - Copy each file to its canonical location
- Append the AGENTS.md entry from the packet
- Run CI validation:
bash scripts/lint-skills-frontmatter.sh && bash scripts/validate-agents-md.sh && bash scripts/validate-commands.sh - If validation passes, delete the staging folder:
_staging/pm-skill-builder/{skill-name}/ - If validation fails, report the error and keep staging intact for fixes
Design rationale lives in the GitHub issue, PR, or effort brief . not
in a permanent packet file.
Provide post-promotion guidance:
- "Run CI locally: "
bash scripts/lint-skills-frontmatter.sh - "Test the skill: try with a realistic scenario"
/{command-name} - "If contributing: create a GitHub issue with the skill-proposal template, then open a PR"
询问:“请审核以上实现包。准备就绪后,我会将文件推广到正式位置。是否继续?[是/否]”
若用户同意,将文件从暂存区复制到正式位置:
| 暂存文件 | 正式位置 |
|---|---|
| |
| |
| |
| |
其中是带分类前缀的目录(例如:)。
{dir-name}deliver-change-communication然后执行以下操作:
- 创建目标目录:
skills/{dir-name}/references/ - 将每个文件复制到其正式位置
- 将实现包中的AGENTS.md条目追加到对应文件
- 运行CI验证:
bash scripts/lint-skills-frontmatter.sh && bash scripts/validate-agents-md.sh && bash scripts/validate-commands.sh - 若验证通过,删除暂存文件夹:
_staging/pm-skill-builder/{skill-name}/ - 若验证失败,报告错误并保留暂存区以便修复
设计依据记录在GitHub issue、PR或工作简报中——不存储在永久的实现包文件中。
提供推广后的指导:
- “本地运行CI:”
bash scripts/lint-skills-frontmatter.sh - “测试技能:尝试使用执行真实场景”
/{command-name} - “若贡献代码:使用skill-proposal模板创建GitHub issue,然后提交PR”
Current Library Reference
当前库参考表
Use this table for gap analysis . it reflects the current skill inventory
(38 skills). Also scan the directory for the latest count.
skills/使用此表进行差距分析——它反映了当前的技能清单(38个技能)。同时请扫描目录获取最新数量。
skills/Domain Skills (25)
领域技能(25个)
| Phase | Skill | Category | Description |
|---|---|---|---|
| discover | competitive-analysis | research | Structured competitive landscape analysis |
| discover | interview-synthesis | research | User research interview synthesis |
| discover | stakeholder-summary | research | Stakeholder needs and influence mapping |
| define | hypothesis | ideation | Testable hypothesis with success metrics |
| define | jtbd-canvas | problem-framing | Jobs to Be Done canvas |
| define | opportunity-tree | problem-framing | Opportunity solution tree |
| define | problem-statement | problem-framing | Clear problem statement with success criteria |
| develop | adr | specification | Architecture Decision Record |
| develop | design-rationale | specification | Design decision reasoning |
| develop | solution-brief | ideation | One-page solution overview |
| develop | spike-summary | coordination | Technical/design spike results |
| deliver | acceptance-criteria | specification | Given/When/Then acceptance criteria |
| deliver | edge-cases | specification | Edge cases and error states |
| deliver | launch-checklist | coordination | Pre-launch checklist |
| deliver | prd | specification | Product Requirements Document |
| deliver | release-notes | coordination | User-facing release notes |
| deliver | user-stories | specification | User stories with acceptance criteria |
| measure | dashboard-requirements | validation | Analytics dashboard spec |
| measure | experiment-design | validation | A/B test or experiment design |
| measure | experiment-results | reflection | Experiment results and learnings |
| measure | instrumentation-spec | validation | Event tracking specification |
| iterate | lessons-log | reflection | Structured lessons learned |
| iterate | pivot-decision | reflection | Pivot or persevere decision |
| iterate | refinement-notes | coordination | Backlog refinement outcomes |
| iterate | retrospective | reflection | Team retrospective |
| 阶段 | 技能 | 类别 | 描述 |
|---|---|---|---|
| discover | competitive-analysis | research | 结构化竞争格局分析 |
| discover | interview-synthesis | research | 用户研究访谈总结 |
| discover | stakeholder-summary | research | 利益相关者需求与影响力映射 |
| define | hypothesis | ideation | 带成功指标的可测试假设 |
| define | jtbd-canvas | problem-framing | Jobs to Be Done画布 |
| define | opportunity-tree | problem-framing | 机会解决方案树 |
| define | problem-statement | problem-framing | 明确的问题陈述及成功标准 |
| develop | adr | specification | 架构决策记录 |
| develop | design-rationale | specification | 设计决策理由 |
| develop | solution-brief | ideation | 单页解决方案概述 |
| develop | spike-summary | coordination | 技术/设计探索结果 |
| deliver | acceptance-criteria | specification | Given/When/Then验收标准 |
| deliver | edge-cases | specification | 边缘情况与错误状态 |
| deliver | launch-checklist | coordination | 上线前检查表 |
| deliver | prd | specification | 产品需求文档 |
| deliver | release-notes | coordination | 用户面向的发布说明 |
| deliver | user-stories | specification | 带验收标准的用户故事 |
| measure | dashboard-requirements | validation | 分析仪表盘规格 |
| measure | experiment-design | validation | A/B测试或实验设计 |
| measure | experiment-results | reflection | 实验结果与经验总结 |
| measure | instrumentation-spec | validation | 事件追踪规格 |
| iterate | lessons-log | reflection | 结构化经验教训记录 |
| iterate | pivot-decision | reflection | 转型或坚持决策 |
| iterate | refinement-notes | coordination | 待办事项梳理成果 |
| iterate | retrospective | reflection | 团队回顾 |
Foundation Skills (1)
基础技能(1个)
| Skill | Category | Description |
|---|---|---|
| persona | research | Evidence-calibrated product or marketing persona |
| 技能 | 类别 | 描述 |
|---|---|---|
| persona | research | 基于证据校准的产品或营销用户画像 |
Utility Skills (1)
实用技能(1个)
| Skill | Category | Description |
|---|---|---|
| pm-skill-builder | coordination | This skill |
| 技能 | 类别 | 描述 |
|---|---|---|
| pm-skill-builder | coordination | 本技能 |
Output Contract
输出契约
The builder MUST produce draft files for the new skill:
- . full skill instructions
SKILL.md - . output template with guidance comments
references/TEMPLATE.md - . complete worked example (150-300 lines)
references/EXAMPLE.md - . slash command file
command.md
All drafts are written to (gitignored).
_staging/pm-skill-builder/{skill-name}/On promotion, files are copied to canonical locations, AGENTS.md is
updated, and the staging folder is discarded.
构建器必须为新技能生成以下草稿文件:
- ——完整的技能说明
SKILL.md - ——带指导注释的输出模板
references/TEMPLATE.md - ——完整的示例(150-300行)
references/EXAMPLE.md - ——斜杠命令文件
command.md
所有草稿均写入(已被git忽略)。
_staging/pm-skill-builder/{skill-name}/推广时,文件将被复制到正式位置,AGENTS.md会更新,暂存文件夹将被删除。
Quality Checklist
质量检查表
Before finalizing the packet, verify all items in both tiers:
在最终确定实现包前,验证以下两个层级的所有项:
CI Validation (must pass)
CI验证(必须通过)
- matches directory name
name - Description is 20-100 words (single-line, no multiline YAML)
- ,
version,updatedall presentlicense - Classification correct (domain → , foundation/utility →
phase:)classification: - Directory name follows convention:
{phase/classification}-{skill-name} - TEMPLATE.md has ≥3 sections
## - Command file references correct skill path
- AGENTS.md entry uses +
####format**Path:**
- 与目录名称匹配
name - 描述为20-100字(单行,无多行YAML)
- 、
version、updated均存在license - 分类正确(领域技能→,基础/实用技能→
phase:)classification: - 目录名称遵循规范:
{phase/classification}-{skill-name} - TEMPLATE.md包含≥3个章节
## - 命令文件引用正确的技能路径
- AGENTS.md条目使用+
####格式**Path:**
Quality Checks (should pass)
质量检查(建议通过)
- Gap analysis checked all existing skills (not just same-phase)
- Why Gate evidence is specific (names prompts/scenarios, not vague)
- EXAMPLE.md is a complete artifact (150-300 lines), not an outline
- Output contract is present in draft SKILL.md
- Quality checklist is present in draft SKILL.md
- Quality Forecast identifies dominant zone and provides writing guidance
- 差距分析检查了所有现有技能(不仅是同阶段)
- Why Gate证据具体(列出提示/场景,而非模糊描述)
- EXAMPLE.md是完整的成果(150-300行),而非大纲
- 输出契约存在于草稿SKILL.md中
- 质量检查表存在于草稿SKILL.md中
- 质量预测明确了主导区域并提供写作指南
Examples
示例
See for a completed Skill Implementation Packet
demonstrating a realistic domain skill creation.
references/EXAMPLE.md查看获取已完成的技能实现包示例,展示真实的领域技能创建流程。
references/EXAMPLE.md