cmd-speckit-tasks

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/speckit.tasks

Agent skill wrapper for the Claude command

/speckit.tasks

When the original command text references

{{INPUT}}

$1

, or named arguments, map them from the user's current request.

Claude命令

/speckit.tasks

的Agent技能封装。

当原始命令文本引用

{{INPUT}}

、

$1

或命名参数时，从用户当前请求中映射对应内容。

Command Instructions

命令说明

Generate a dependency-ordered

tasks.md

with phase-based tasks derived from the feature specification and plan.

基于功能规格说明和计划，生成按依赖顺序排列的分阶段任务文档

tasks.md

。

Steps

步骤

Locate artifacts:
- If input names a feature or directory, use it.
- Otherwise check
```
.specify/specs/
```
  and ask if multiple features exist.
- Load:
```
spec.md
```
  (user stories + priorities),
```
plan.md
```
  (tech stack, phases). Optionally:
```
data-model.md
```
  ,
```
contracts/
```
  ,
```
research.md
```
  .
- Load
```
.specify/memory/constitution.md
```
  .
- Set
```
FEATURE_DIR
```
  to the feature directory.
Extract user stories and priorities: From
```
spec.md
```
, extract each user story. Assign priority:
- P1 = critical path (required for MVP)
- P2 = important (needed for completion)
- P3 = nice-to-have (can defer)

Generate
FEATURE_DIR/tasks.md
:

markdown

# Tasks: <Feature Name>

## Phase 1: Setup
Shared infrastructure and scaffolding required by all subsequent phases.
- [ ] [T001] [P] <task description> — <file path if applicable>
- [ ] [T002] <task description>

## Phase 2: Foundation
Blocking prerequisites that multiple user stories depend on.
- [ ] [T010] [P] <task description>

## Phase 3: <User Story 1 Name> [US1]
Implement [US1]: "As a [role], I want to [action]..."
- [ ] [T020] [P] [US1] Write unit tests for <component>
- [ ] [T021] [US1] Implement <component> core logic

## Phase N: Polish & Cross-Cutting Concerns
- [ ] [TN00] [P] Add error handling across all endpoints
- [ ] [TN01] Write integration tests for full workflow
- [ ] [TN02] Update documentation

Task format rules:

```
[P]
```
marks a task as parallelizable (no blocking dependencies within the phase)
```
[USn]
```
traces the task to user story n (Phase 3+ only)
Task IDs increment sequentially (T001, T002, ...)
Include file paths in task description when the task targets a specific file

Report: Task counts by phase, total parallelizable task count, which phases constitute the MVP scope.

Next step: Run

/speckit.checklist

to validate spec quality, or

/speckit.implement

to begin implementation.

定位工件:
- 如果输入指定了功能或目录，则使用该内容。
- 否则检查
```
.specify/specs/
```
  目录，若存在多个功能则询问用户。
- 加载文件：
```
spec.md
```
  （用户故事+优先级）、
```
plan.md
```
  （技术栈、阶段）。可选加载：
```
data-model.md
```
  、
```
contracts/
```
  、
```
research.md
```
  。
- 加载
```
.specify/memory/constitution.md
```
  。
- 将
```
FEATURE_DIR
```
  设置为功能目录。
提取用户故事和优先级：从
```
spec.md
```
中提取每个用户故事，并分配优先级：
- P1 = 关键路径（MVP必备）
- P2 = 重要需求（完成项目所需）
- P3 = 锦上添花（可延迟实现）

生成
FEATURE_DIR/tasks.md
:

markdown

# Tasks: <Feature Name>

## Phase 1: Setup
Shared infrastructure and scaffolding required by all subsequent phases.
- [ ] [T001] [P] <task description> — <file path if applicable>
- [ ] [T002] <task description>

## Phase 2: Foundation
Blocking prerequisites that multiple user stories depend on.
- [ ] [T010] [P] <task description>

## Phase 3: <User Story 1 Name> [US1]
Implement [US1]: "As a [role], I want to [action]..."
- [ ] [T020] [P] [US1] Write unit tests for <component>
- [ ] [T021] [US1] Implement <component> core logic

## Phase N: Polish & Cross-Cutting Concerns
- [ ] [TN00] [P] Add error handling across all endpoints
- [ ] [TN01] Write integration tests for full workflow
- [ ] [TN02] Update documentation

任务格式规则:

```
[P]
```
标记任务为可并行执行（阶段内无阻塞依赖）
```
[USn]
```
将任务追溯到第n个用户故事（仅第3阶段及以后使用）
任务ID按顺序递增（T001、T002……）
当任务针对特定文件时，在任务描述中包含文件路径

报告：按阶段统计任务数量、可并行任务总数，以及哪些阶段属于MVP范围。

下一步：运行

/speckit.checklist

验证规格说明质量，或运行

/speckit.implement

开始实施。