prd-planner
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChinesePRD Planner
PRD 规划工具
A PRD creation skill that uses persistent file-based planning to maintain coherent thinking and avoid "left-brain vs right-brain" context switching issues.
一款采用持久化文件规划的PRD创建技能,可保持思维连贯性,避免“左脑vs右脑”的上下文切换问题。
When This Skill Activates
该技能的触发条件
This skill activates when you:
- Explicitly say "PRD", "prd", "create a PRD", or "产品需求文档"
- Say "product requirements document" or "产品需求"
- Mention "write a PRD for..."
- Say "PRD planning" or "PRD 设计"
If user says "design solution" or "architecture design" without mentioning PRD, use instead.
architecting-solutions当你出现以下表述时,该技能会被激活:
- 明确提及“PRD”、“prd”、“create a PRD”或“产品需求文档”
- 提及“product requirements document”或“产品需求”
- 提到“write a PRD for...”
- 提及“PRD planning”或“PRD 设计”
如果用户仅说“design solution”或“architecture design”但未提及PRD,请使用技能。
architecting-solutionsThe Core Philosophy
核心理念
"PRD creation should be traceable, coherent, and persistent - not scattered across context switches."
This skill combines:
- PRD methodology (from architecting-solutions)
- File-based persistence (from planning-with-files)
To create a single, coherent PRD creation workflow that doesn't lose context.
“PRD的创建应具备可追溯性、连贯性和持久性——不应因上下文切换而分散。”
本技能结合了:
- PRD方法论(源自architecting-solutions)
- 基于文件的持久化机制(源自planning-with-files)
打造单一、连贯的PRD创建工作流,避免上下文丢失。
4-File Pattern for PRD Creation
PRD创建的四文件模式
For every PRD project, create FOUR files:
Pick a SCOPE (short, unique, kebab-case slug) and use it as a prefix for all files.
text
docs/{scope}-prd-notes.md → Store research, requirements, findings, options
docs/{scope}-prd-task-plan.md → Track PRD creation phases and progress
docs/{scope}-prd.md → Product requirements (what & why)
docs/{scope}-tech.md → Technical design (how)每个PRD项目都需要创建四个文件:
选择一个SCOPE(简短、唯一的短横线命名格式标识),并将其作为所有文件的前缀。
text
docs/{scope}-prd-notes.md → 存储调研内容、需求、发现的问题、可选方案
docs/{scope}-prd-task-plan.md → 跟踪PRD创建的阶段与进度
docs/{scope}-prd.md → 产品需求(做什么&为什么做)
docs/{scope}-tech.md → 技术设计(怎么做)File Purposes
文件用途
| File | Purpose | Audience | Updated When |
|---|---|---|---|
| Raw research, requirements, architecture options (A/B/C) | Self + reviewers | New information gathered |
| Track progress, phases, checkboxes, timestamps | PM + dev lead | Each phase completion |
| Product requirements (what & why), user flows | PM + stakeholders + devs | After requirements are clear |
| Technical design (API, data flow, implementation) | Developers + architects | After architecture is decided |
| 文件 | 用途 | 受众 | 更新时机 |
|---|---|---|---|
| 原始调研、需求、架构可选方案(A/B/C) | 自身 + 评审人员 | 收集到新信息时 |
| 跟踪进度、阶段、复选框、时间戳 | 产品经理 + 开发负责人 | 每个阶段完成后 |
| 产品需求(做什么&为什么做)、用户流程 | 产品经理 + 利益相关者 + 开发人员 | 需求明确后 |
| 技术设计(API、数据流、实现方案) | 开发人员 + 架构师 | 架构方案确定后 |
Workflow
工作流
┌─────────────────────────────────────────────────────────────────┐
│ PRD Creation Workflow │
├─────────────────────────────────────────────────────────────────┤
│ │
│ 1. Initialize → Create 4 files with template │
│ 2. Requirements → Gather to {scope}-prd-notes.md │
│ 2.5 Edge Cases → Scan codebase, infer patterns, ask smartly │
│ 3. Analysis → Research best practices, save to notes │
│ 4. Design → Propose architecture options (A/B/C), save to notes │
│ 5. PRD → Write product requirements to {scope}-prd.md │
│ 6. Tech → Write technical design to {scope}-tech.md │
│ 7. Validate → Review with user, finalize │
│ │
└─────────────────────────────────────────────────────────────────┘
↓
All thinking persisted to files
↓
No context switching┌─────────────────────────────────────────────────────────────────┐
│ PRD 创建工作流 │
├─────────────────────────────────────────────────────────────────┤
│ │
│ 1. 初始化 → 使用模板创建4个文件 │
│ 2. 需求收集 → 将内容存入{scope}-prd-notes.md │
│ 2.5 边缘场景分析 → 扫描代码库、推导模式、精准提问 │
│ 3. 分析调研 → 调研最佳实践,存入笔记文件 │
│ 4. 方案设计 → 提出架构可选方案(A/B/C),存入笔记文件 │
│ 5. 撰写PRD → 将产品需求写入{scope}-prd.md │
│ 6. 撰写技术文档 → 将技术设计写入{scope}-tech.md │
│ 7. 验证定稿 → 与用户评审、最终确认 │
│ │
└─────────────────────────────────────────────────────────────────┘
↓
所有思考内容持久化存储到文件
↓
无上下文切换Step 1: Initialize
步骤1:初始化
Create the four files with templates:
使用模板创建四个文件:
{scope}-prd-task-plan.md
{scope}-prd-task-plan.md
markdown
undefinedmarkdown
undefinedPRD Task Plan: {Feature Name}
PRD 任务计划: {功能名称}
Goal
目标
Create a PRD and technical design for {feature description}.
为{功能描述}创建PRD与技术设计文档。
Owner
负责人
{User name/role}
{用户姓名/角色}
Phases
阶段
- Phase 1: Initialize files ✓
- Phase 2: Gather requirements (CURRENT)
- Phase 3: Research & analysis
- Phase 4: Design solution
- Phase 5: Write PRD
- Phase 6: Write technical design
- Phase 7: Validate & finalize
- 阶段1:初始化文件 ✓
- 阶段2:收集需求(当前阶段)
- 阶段3:调研与分析
- 阶段4:设计解决方案
- 阶段5:撰写PRD
- 阶段6:撰写技术设计文档
- 阶段7:验证与定稿
Status
状态
Currently in Phase 2 - Gathering requirements from user
当前处于阶段2 - 正在从用户处收集需求
Progress Log
进度日志
- {timestamp} - Phase 1 complete: Files initialized
undefined- {时间戳} - 阶段1完成:文件初始化完成
undefined{scope}-prd-notes.md
{scope}-prd-notes.md
markdown
undefinedmarkdown
undefinedPRD Notes: {Feature Name}
PRD 笔记: {功能名称}
Raw Requirements
原始需求
(Add user requirements as they emerge)
(随需求浮现逐步添加)
Constraints
约束条件
(Add technical, business, time constraints)
(添加技术、业务、时间方面的约束)
Inferred Patterns (from codebase)
推导的模式(来自代码库)
| Edge Case | Source | Pattern Applied |
|---|---|---|
| (Filled after Step 2.5 codebase scan) |
| 边缘场景 | 来源 | 采用的模式 |
|---|---|---|
| (步骤2.5扫描代码库后填充) |
Edge Cases
边缘场景
Auto-handled (following codebase patterns)
自动处理(遵循代码库模式)
- (Filled after Step 2.5 analysis)
- (步骤2.5分析后填充)
Confirmed by User
用户确认
- (Filled after user confirms edge case decisions)
- (用户确认边缘场景决策后填充)
Open Questions
待确认问题
- (Track questions to ask user)
- (跟踪需要向用户提出的问题)
Research Findings
调研结果
(Add research on best practices, similar solutions)
(添加最佳实践、同类方案的调研内容)
Architecture Options
架构可选方案
-
Option A: {Description}
- Pros: {Advantages}
- Cons: {Disadvantages}
-
Option B: {Description}
- Pros: {Advantages}
- Cons: {Disadvantages}
-
Option C: {Description}
- Pros: {Advantages}
- Cons: {Disadvantages}
Selected: Option {X}
undefined-
方案A: {描述}
- 优势: {优点}
- 劣势: {缺点}
-
方案B: {描述}
- 优势: {优点}
- 劣势: {缺点}
-
方案C: {描述}
- 优势: {优点}
- 劣势: {缺点}
选定方案: 方案{X}
undefined{scope}-prd.md
{scope}-prd.md
markdown
undefinedmarkdown
undefinedPRD: {Feature Name}
PRD: {功能名称}
Status: DRAFT Last updated: {timestamp}
状态: 草稿 最后更新: {时间戳}
Table of Contents
目录
Problem Statement
问题描述
To be filled after requirements gathering
需求收集完成后填充
Goals and Non-Goals
目标与非目标
Goals
目标
- {Specific achievable outcomes}
- {具体可实现的成果}
Non-Goals
非目标
- {Explicit exclusions}
- {明确排除的内容}
Success Criteria
成功标准
To be filled with measurable criteria
填充可衡量的标准
Scope
范围
In Scope
包含范围
- {Specific items included}
- {具体包含的内容}
Out of Scope
排除范围
- {Specific items excluded}
... (rest of PRD sections)
undefined- {具体排除的内容}
...(PRD其余部分)
undefined{scope}-tech.md
{scope}-tech.md
markdown
undefinedmarkdown
undefinedTechnical Design: {Feature Name}
技术设计: {功能名称}
Status: DRAFT Last updated: {timestamp}
状态: 草稿 最后更新: {时间戳}
Overview
概述
{High-level technical approach}
{技术方案的高层级说明}
Key Components
核心组件
{List major components and their responsibilities}
{列出主要组件及其职责}
API Design
API设计
{API signatures, request/response formats}
{API签名、请求/响应格式}
Data Flow
数据流
{How data flows through the system}
{数据在系统中的流转方式}
Implementation Details
实现细节
{Specific implementation notes}
{具体的实现说明}
Migration Plan
迁移计划
{If applicable, how to migrate from existing system}
undefined{若适用,说明从现有系统迁移的方式}
undefinedStep 2: Gather Requirements
步骤2:收集需求
Ask clarifying questions and save responses to :
{scope}-prd-notes.md提出澄清问题并将回复存入:
{scope}-prd-notes.mdCore Questions to Ask
核心提问
- Problem: What problem are we solving?
- Users: Who will use this?
- Success: How do we know it's successful?
- Constraints: Any technical/time/budget constraints?
Save each answer to under appropriate section.
{scope}-prd-notes.mdAlways update after gathering info:
{scope}-prd-task-plan.mdmarkdown
- [x] Phase 2: Gather requirements ✓
- [ ] Phase 2.5: Edge case analysis (CURRENT)
- [ ] Phase 3: Research & analysis- 问题: 我们要解决什么问题?
- 用户: 谁会使用该功能?
- 成功: 如何判断功能成功?
- 约束: 是否有技术/时间/预算方面的约束?
将每个答案存入对应的章节。
{scope}-prd-notes.md收集信息后务必更新:
{scope}-prd-task-plan.mdmarkdown
- [x] 阶段2:收集需求 ✓
- [ ] 阶段2.5:边缘场景分析(当前阶段)
- [ ] 阶段3:调研与分析Step 2.5: Context-Aware Edge Case Analysis
步骤2.5:上下文感知的边缘场景分析
Before asking users about edge cases, scan the codebase first to infer existing patterns. This reduces redundant questions and ensures consistency with the project.
Detailed reference: Seefor full scanning commands and output formats.references/edge-case-analysis.md
在向用户询问边缘场景之前,先扫描代码库推导现有模式。这可减少重复提问,确保与项目保持一致。
详细参考: 完整的扫描命令与输出格式请查看。references/edge-case-analysis.md
Quick Process
快速流程
- Scan codebase for existing patterns (delete strategy, error handling, empty states, pagination)
- Identify requirement type (CRUD, State Workflow, Async, Data Display, Form, File)
- Generate smart assumptions - patterns found in code don't need user confirmation
- Ask only when needed - no precedent, multiple patterns, or business decision required
- 扫描代码库寻找现有模式(删除策略、错误处理、空状态、分页)
- 识别需求类型(CRUD、状态工作流、异步处理、数据展示、表单、文件)
- 生成智能假设 - 代码中已有的模式无需用户确认
- 仅在必要时提问 - 无先例、存在多种模式或需要业务决策时才提问
When to Ask Users
何时需要询问用户
| Condition | Action |
|---|---|
| Pattern exists in codebase | Auto-apply, no question needed |
| No precedent found | Ask user with options |
| Multiple conflicting patterns | Ask user to choose |
| Business rule required | Ask user |
| 条件 | 操作 |
|---|---|
| 代码库中存在对应模式 | 自动应用,无需提问 |
| 无先例可循 | 向用户提供选项并提问 |
| 存在多种冲突模式 | 请用户选择 |
| 需要业务规则决策 | 向用户提问 |
Output to Notes File
存入笔记文件
Update with:
{scope}-prd-notes.mdmarkdown
undefined更新:
{scope}-prd-notes.mdmarkdown
undefinedInferred Patterns (from codebase)
推导的模式(来自代码库)
| Edge Case | Source | Pattern Applied |
|---|---|---|
| Delete | | Soft delete |
| 边缘场景 | 来源 | 采用的模式 |
|---|---|---|
| 删除 | | 软删除 |
Edge Cases
边缘场景
Auto-handled (following codebase patterns)
自动处理(遵循代码库模式)
- Empty list → Use existing EmptyState component
- 空列表 → 使用现有EmptyState组件
Confirmed by User
用户确认
- Concurrent edit: Last write wins (confirmed {date})
**Update task plan:**
```markdown
- [x] Phase 2.5: Edge case analysis ✓
- [ ] Phase 3: Research & analysis (CURRENT)- 并发编辑:最后写入者获胜(确认于{日期})
**更新任务计划:**
```markdown
- [x] 阶段2.5:边缘场景分析 ✓
- [ ] 阶段3:调研与分析(当前阶段)Step 3: Research & Analysis
步骤3:调研与分析
Research best practices and save to :
{scope}-prd-notes.mdbash
undefined调研最佳实践并存入:
{scope}-prd-notes.mdbash
undefinedSearch for similar implementations
搜索同类实现
grep -r "keyword" packages/ --include="*.ts"
grep -r "keyword" packages/ --include="*.ts"
Search web for best practices
搜索业内最佳实践
web search "best practices for {feature}"
Save findings to `{scope}-prd-notes.md` → Research Findings section.web search "best practices for {feature}"
将调研结果存入`{scope}-prd-notes.md`的「调研结果」章节。Step 4: Design Solution
步骤4:设计解决方案
Propose architecture with trade-offs, save to :
{scope}-prd-notes.mdmarkdown
undefined提出包含权衡的架构方案,存入:
{scope}-prd-notes.mdmarkdown
undefinedArchitecture Options
架构可选方案
-
Option A: {Description}
- Pros: {Advantages}
- Cons: {Disadvantages}
-
Option B: {Description}
- Pros: {Advantages}
- Cons: {Disadvantages}
-
Option C: {Description}
- Pros: {Advantages}
- Cons: {Disadvantages}
Selected: Option {X} - because {reason}
undefined-
方案A: {描述}
- 优势: {优点}
- 劣势: {缺点}
-
方案B: {描述}
- 优势: {优点}
- 劣势: {缺点}
-
方案C: {描述}
- 优势: {优点}
- 劣势: {缺点}
选定方案: 方案{X} - 原因:{理由}
undefinedStep 5: Write PRD
步骤5:撰写PRD
Read and synthesize into polished PRD:
{scope}-prd-notes.mdmarkdown
1. Read {scope}-prd-notes.md to understand:
- Requirements gathered
- Research findings
- Architecture decision (which option was selected)
2. Write {scope}-prd.md with:
- Clear problem statement
- Goals and Non-Goals (explicit exclusions)
- Measurable success criteria (specific numbers/timings)
- Scope (In Scope / Out of Scope)
- Functional requirements
- Non-functional requirements
- User flows
- Implementation plan (high level)
3. Reference tech doc: "See {scope}-tech.md for technical design"阅读并整理为规范的PRD文档:
{scope}-prd-notes.mdmarkdown
1. 阅读{scope}-prd-notes.md,理解:
- 已收集的需求
- 调研结果
- 架构决策(选定的方案)
2. 撰写{scope}-prd.md,包含:
- 清晰的问题描述
- 目标与非目标(明确排除项)
- 可衡量的成功标准(具体数字/时间)
- 范围(包含/排除)
- 功能需求
- 非功能需求
- 用户流程
- 实施计划(高层级)
3. 引用技术文档:“技术设计请查看{scope}-tech.md”Step 6: Write Technical Design
步骤6:撰写技术设计文档
markdown
1. Read {scope}-prd-notes.md for selected architecture option
2. Write {scope}-tech.md with:
- Overview (technical approach summary)
- Key Components (what pieces, responsibilities)
- API Design (signatures, contracts)
- Data Flow (how data moves through system)
- Implementation Details (specific notes)
- Migration Plan (if applicable)markdown
1. 阅读{scope}-prd-notes.md中选定的架构方案
2. 撰写{scope}-tech.md,包含:
- 概述(技术方案总结)
- 核心组件(组成部分及职责)
- API设计(签名、契约)
- 数据流(数据在系统中的流转方式)
- 实现细节(具体说明)
- 迁移计划(若适用)Step 7: Validate & Finalize
步骤7:验证与定稿
Review with user:
- Present PRD summary
- Ask for feedback
- Incorporate changes
- Mark Phase 7 complete
与用户评审:
- 展示PRD摘要
- 征求反馈
- 整合修改意见
- 标记阶段7完成
Important Rules
重要规则
| Rule | Bad | Good |
|---|---|---|
| Use Files | Keep in memory | Save to {scope}-prd-notes.md |
| Update Plan | Move on without update | Update task-plan.md with checkbox |
| Read Before Decide | Decide from memory | Read notes first |
| Separate Docs | Mix PRD + Tech | PRD for "what", Tech for "how" |
| Include Options | Jump to solution | Document 2-3 options with pros/cons |
| 规则 | 错误做法 | 正确做法 |
|---|---|---|
| 使用文件存储 | 仅保存在内存中 | 存入{scope}-prd-notes.md |
| 更新任务计划 | 直接进入下一阶段不更新 | 更新task-plan.md的复选框 |
| 先阅读再决策 | 凭记忆做决策 | 先阅读笔记内容 |
| 分离文档 | 混合PRD与技术内容 | PRD聚焦“做什么”,技术文档聚焦“怎么做” |
| 提供可选方案 | 直接给出单一解决方案 | 记录2-3个带优缺点的可选方案 |
Phase Transitions
阶段过渡
Update after each phase with checkbox ✓ and timestamp.
{scope}-prd-task-plan.md每个阶段完成后,更新的复选框为✓并添加时间戳。
{scope}-prd-task-plan.mdCompleting a PRD
PRD完成
Mark all phases complete, set status to "✅ COMPLETE", log final deliverables.
标记所有阶段为完成,将状态设置为“✅ 已完成”,记录最终交付物。
File Cleanup (Optional)
文件清理(可选)
After PRD is complete:
- Keep for reference (shows decision process)
{scope}-prd-notes.md - Archive or delete
{scope}-prd-task-plan.md - Final outputs are and
{scope}-prd.md{scope}-tech.md
PRD完成后:
- 保留供参考(展示决策过程)
{scope}-prd-notes.md - 归档或删除
{scope}-prd-task-plan.md - 最终输出为和
{scope}-prd.md{scope}-tech.md
Quick Start Template
快速启动模板
markdown
undefinedmarkdown
undefinedPRD Task Plan: {Feature}
PRD 任务计划: {功能}
Goal
目标
Create PRD and technical design for {description}
为{描述}创建PRD与技术设计文档
Phases
阶段
- Initialize 4 files
- Gather requirements
- Research & analysis
- Design solution (A/B/C options)
- Write PRD
- Write technical design
- Validate & finalize
- 初始化4个文件
- 收集需求
- 调研与分析
- 设计解决方案(A/B/C可选方案)
- 撰写PRD
- 撰写技术设计文档
- 验证与定稿
Status
状态
Phase 1: Initializing files
undefined阶段1:初始化文件
undefinedWhy This Works
为何该方案有效
| Problem | Solution |
|---|---|
| Context switching | All thinking in files, read anytime |
| Lost requirements | Saved to {scope}-prd-notes.md immediately |
| Inconsistent PRDs | Same process, same structure |
| "Left brain vs right brain" | One coherent workflow |
| Re-explaining context | Files contain full context |
| Mixed concerns | PRD (product) separate from Tech (implementation) |
| Hidden decisions | Architecture options A/B/C documented |
| 问题 | 解决方案 |
|---|---|
| 上下文切换 | 所有思考内容存入文件,可随时查阅 |
| 需求丢失 | 立即存入{scope}-prd-notes.md |
| PRD不一致 | 统一流程与结构 |
| “左脑vs右脑”切换 | 单一连贯的工作流 |
| 重复解释上下文 | 文件包含完整上下文 |
| 关注点混合 | PRD(产品)与技术文档(实现)分离 |
| 决策不透明 | 记录A/B/C架构可选方案 |
References
参考资料
- planning-with-files - File-based planning methodology
- architecting-solutions - PRD creation best practices
- Edge case scanning:
references/edge-case-analysis.md
- planning-with-files - 基于文件的规划方法论
- architecting-solutions - PRD创建最佳实践
- 边缘场景扫描:
references/edge-case-analysis.md
Auto-Trigger (Automation)
自动触发(自动化)
When this skill completes, automatically trigger:
- self-improving-agent (background) - Extract patterns
- session-logger (auto) - Save session context
该技能完成后,自动触发:
- self-improving-agent(后台)- 提取模式
- session-logger(自动)- 保存会话上下文