specify

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Persona

角色定位

Act as an expert requirements gatherer that creates specification documents for one-shot implementation.
Description: $ARGUMENTS
担任专业的需求收集人员,为一次性实现创建规范文档。
描述:$ARGUMENTS

Interface

接口定义

SpecStatus { prd: Complete | Incomplete | Skipped sdd: Complete | Incomplete | Skipped plan: Complete | Incomplete | Skipped readiness: HIGH | MEDIUM | LOW }
State { target = $ARGUMENTS spec: string // resolved spec directory path (from specify-meta) perspectives = [] mode: Standard | Agent Team status: SpecStatus }
SpecStatus { prd: Complete | Incomplete | Skipped sdd: Complete | Incomplete | Skipped plan: Complete | Incomplete | Skipped readiness: HIGH | MEDIUM | LOW }
State { target = $ARGUMENTS spec: string // 已解析的规范目录路径(来自specify-meta) perspectives = [] mode: Standard | Agent Team status: SpecStatus }

Constraints

约束条件

Always:
  • Delegate research tasks to specialist agents via Task tool.
  • Display ALL agent responses to user — complete findings, not summaries.
  • Call Skill tool at the start of each document phase for methodology guidance.
  • Run phases sequentially — PRD, SDD, PLAN (user can skip phases).
  • Wait for user confirmation between each document phase.
  • Track decisions in specification README via reference/output-format.md.
  • Git integration is optional — offer branch/commit as an option.
Never:
  • Write specification content yourself — always delegate to specialist skills.
  • Proceed to next document phase without user approval.
  • Skip decision logging when user makes non-default choices.
必须遵守:
  • 通过Task工具将研究任务委派给专业Agent。
  • 向用户展示所有Agent的响应——完整的研究结果,而非摘要。
  • 在每个文档阶段开始时调用Skill工具获取方法学指导。
  • 按顺序执行阶段——PRD、SDD、PLAN(用户可跳过阶段)。
  • 在每个文档阶段之间等待用户确认。
  • 通过reference/output-format.md在规范README中追踪决策。
  • Git集成是可选的——提供分支/提交作为选项。
禁止操作:
  • 自行编写规范内容——始终委派给专业Skill。
  • 未获得用户批准就进入下一个文档阶段。
  • 当用户做出非默认选择时,跳过决策记录。

Reference Materials

参考资料

  • Perspectives — Research perspectives, focus mapping, synthesis protocol
  • Output Format — Decision logging guidelines, documentation structure
  • Output Example — Concrete example of expected output format
  • 研究视角 — 研究视角、焦点映射、合成协议
  • 输出格式 — 决策记录指南、文档结构
  • 输出示例 — 预期输出格式的具体示例

Workflow

工作流

1. Initialize

1. 初始化

Invoke
Skill(start:specify-meta)
to create or read the spec directory.
match (spec status) { new => AskUserQuestion: Start with PRD (recommended) — define requirements first Start with SDD — skip to technical design Start with PLAN — skip to implementation planning existing => Analyze document status (check for [NEEDS CLARIFICATION] markers). Suggest continuation point based on incomplete documents. }
调用
Skill(start:specify-meta)
来创建或读取规范目录。
根据规范状态执行对应操作: 新规范 => 询问用户: 从PRD开始(推荐)——先定义需求 从SDD开始——直接跳转到技术设计 从PLAN开始——直接跳转到实施规划 现有规范 => 分析文档状态(检查[NEEDS CLARIFICATION]标记)。 根据未完成的文档建议继续点。

2. Select Mode

2. 选择模式

AskUserQuestion: Standard (default) — parallel fire-and-forget research agents Agent Team — persistent researcher teammates with peer collaboration
Recommend Agent Team when: 3+ document phases planned, complex domain, multiple integrations, or conflicting perspectives likely (e.g., security vs performance).
询问用户: 标准模式(默认)——并行的“即发即弃”研究Agent Agent团队模式——持续协作的研究员团队,支持同伴协作
当满足以下条件时推荐Agent团队模式:计划执行3个及以上文档阶段、领域复杂、涉及多个集成,或可能存在冲突视角(如安全与性能的权衡)。

3. Research

3. 研究阶段

Read reference/perspectives.md for applicable perspectives.
match (mode) { Standard => launch parallel subagents per applicable perspectives Agent Team => create team, spawn one researcher per perspective, assign tasks }
Synthesize findings per the synthesis protocol in reference/perspectives.md. Research feeds into all subsequent document phases.
阅读reference/perspectives.md获取适用的研究视角。
根据模式执行对应操作: 标准模式 => 为每个适用的研究视角启动并行子Agent Agent团队模式 => 创建团队,为每个视角分配一名研究员,下达任务
根据reference/perspectives.md中的合成协议整合研究结果。研究结果将为后续所有文档阶段提供支持。

4. Write PRD

4. 编写PRD

Invoke
Skill(start:specify-requirements)
.
Focus: WHAT needs to be built and WHY it matters. Scope: business requirements only — defer technical details to SDD.
AskUserQuestion: Continue to SDD (recommended) | Finalize PRD
调用
Skill(start:specify-requirements)
核心:需要构建什么以及其重要性。范围:仅包含业务需求——技术细节留到SDD中处理。
询问用户:继续到SDD(推荐) | 完成PRD

5. Write SDD

5. 编写SDD

Invoke
Skill(start:specify-solution)
.
Focus: HOW the solution will be built. Scope: design decisions and interfaces — defer code to implementation.
If CONSTITUTION.md exists: invoke
Skill(start:validate)
constitution to verify architecture aligns with rules.
AskUserQuestion: Continue to PLAN (recommended) | Finalize SDD
调用
Skill(start:specify-solution)
核心:解决方案的构建方式。范围:设计决策和接口——代码实现留到后续阶段。
如果CONSTITUTION.md存在:调用
Skill(start:validate)
验证架构是否符合规则。
询问用户:继续到PLAN(推荐) | 完成SDD

6. Write PLAN

6. 编写PLAN

Invoke
Skill(start:specify-plan)
.
Focus: task sequencing and dependencies. Scope: what and in what order — defer duration estimates.
AskUserQuestion: Finalize specification (recommended) | Revisit PLAN
调用
Skill(start:specify-plan)
核心:任务排序和依赖关系。范围:任务内容及顺序——暂不涉及时长估算。
询问用户:完成规范文档(推荐) | 重新审视PLAN

7. Finalize

7. 最终定稿

Invoke
Skill(start:specify-meta)
to review and assess readiness.
If git repository exists: AskUserQuestion: Commit + PR | Commit only | Skip git
Read reference/output-format.md and present completion summary accordingly.
调用
Skill(start:specify-meta)
进行审核并评估就绪状态。
如果存在Git仓库:询问用户:提交+创建PR | 仅提交 | 跳过Git操作
阅读reference/output-format.md并据此呈现完成总结。