cmd-speckit-implement

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/speckit.implement

Agent skill wrapper for the Claude command

/speckit.implement

When the original command text references

{{INPUT}}

$1

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

Claude命令

/speckit.implement

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

Execute the implementation plan for a feature by working through

tasks.md

phase by phase, using TDD where applicable.

通过逐步执行

tasks.md

中的内容来完成功能的实现计划，适用情况下采用TDD方式。

Steps

步骤

Locate artifacts:
- If input names a feature, directory, or phase number, parse it.
- Load required:
```
tasks.md
```
  ,
```
plan.md
```
  . Load optional:
```
data-model.md
```
  ,
```
contracts/
```
  ,
```
research.md
```
  .
- Load
```
.specify/memory/constitution.md
```
  .
- Set
```
FEATURE_DIR
```
  to the feature directory.
Check checklists: Scan
```
FEATURE_DIR/checklists/
```
for any uncompleted items (
```
- [ ]
```
). If found, report them and ask: "There are incomplete checklist items. Proceed anyway?" Stop if the user declines.
Determine starting phase:
- If a phase number was provided in input, start there.
- Otherwise start from Phase 1.
- Skip phases where all tasks are already marked
```
[x]
```
  .
Execute each phase:
- Announce the phase name and task count before starting.
- For tasks marked
```
[P]
```
  (parallelizable): execute them concurrently.
- For tasks without
```
[P]
```
  : execute in listed order.
- For each task:
  - Write tests first (TDD), then implement the minimum code to satisfy the task.
  - Mark the task
```
[x]
```
    in
```
tasks.md
```
    immediately after completion.
  - Run available test and lint commands to verify no regressions.
- If a non-parallelizable task fails, stop and report. Do not proceed to the next task until resolved.
Validate completion after all phases:
- All tasks in
```
tasks.md
```
  are marked
```
[x]
```
  .
- All functional requirements from
```
spec.md
```
  are addressed.
- Tests pass and no new lint errors.
Report: Tasks completed, any skipped tasks with reasons, test results summary.

Next step: Run

/speckit.taskstoissues

to track remaining work as GitHub issues, or commit and open a PR.

定位工件:
- 如果输入指定了功能、目录或阶段编号，解析该输入。
- 加载必需文件：
```
tasks.md
```
  、
```
plan.md
```
  。加载可选文件：
```
data-model.md
```
  、
```
contracts/
```
  、
```
research.md
```
  。
- 加载
```
.specify/memory/constitution.md
```
  。
- 设置
```
FEATURE_DIR
```
  为功能目录。
检查检查清单: 扫描
```
FEATURE_DIR/checklists/
```
中的未完成项（
```
- [ ]
```
）。如果发现未完成项，报告并询问："存在未完成的检查清单项。是否仍要继续？"若用户拒绝则停止执行。
确定起始阶段:
- 如果输入中提供了阶段编号，从该阶段开始。
- 否则从第1阶段开始。
- 跳过所有任务已标记为
```
[x]
```
  的阶段。
执行每个阶段:
- 开始前宣布阶段名称和任务数量。
- 对于标记为
```
[P]
```
  （可并行执行）的任务：并发执行。
- 对于未标记
```
[P]
```
  的任务：按列出的顺序执行。
- 针对每个任务:
  - 先编写测试（TDD），然后实现满足任务要求的最简代码。
  - 任务完成后立即在
```
tasks.md
```
    中将其标记为
```
[x]
```
    。
  - 运行可用的测试和代码检查命令，验证没有回归问题。
- 如果某个非并行任务执行失败，停止并报告。问题解决前不要继续执行下一个任务。
完成后验证（所有阶段完成后）：
- ```
tasks.md
```
  中的所有任务均标记为
```
[x]
```
  。
- ```
spec.md
```
  中的所有功能需求均已处理。
- 测试通过且无新的代码检查错误。
报告: 已完成的任务、任何跳过的任务及原因、测试结果摘要。

下一步: 运行

/speckit.taskstoissues

将剩余工作跟踪为GitHub issues，或提交代码并打开PR。