Back to Details

minimal-run-and-audit

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

minimal-run-and-audit

minimal-run-and-audit

When to apply

适用场景

After a reproduction target and setup plan exist.
When the main skill needs execution evidence and normalized outputs.
When a smoke test, documented inference run, documented evaluation run, or training startup verification is appropriate.
When the user already knows what command should be attempted and wants execution plus reporting only.

已确定复现目标和搭建方案后
主技能需要执行证据和标准化输出时
适合执行冒烟测试、文档记载的推理运行、文档记载的评估运行或训练启动验证时
用户已明确需要尝试的命令，仅需要执行和报告能力时

When not to apply

不适用场景

During initial repo scanning.
When environment or assets are still undefined enough to make execution meaningless.
When the task is a literature lookup rather than repository execution.
When the user is still deciding which reproduction target should count as the main run.

初始仓库扫描阶段
环境或资产仍未明确定义，执行无意义时
任务为文献检索而非仓库执行时
用户仍在确定将哪个复现目标作为主运行项时

Clear boundaries

明确边界

This skill owns normalized reporting for an attempted command.
It may receive execution evidence from the main skill or a thin helper.
It does not choose the overall target on its own.
It does not perform broad paper analysis.
It should not normalize risky code edits into acceptable practice.

本技能负责为尝试执行的命令生成标准化报告
可接收来自主技能或轻量辅助工具的执行证据
不会自主选择整体目标
不执行广泛的论文分析
不应将风险代码编辑规范为可接受实践

Input expectations

输入要求

selected reproduction goal
runnable commands or smoke commands
environment and asset assumptions
optional patch metadata

选定的复现目标
可运行命令或冒烟测试命令
环境和资产假设
可选的补丁元数据

Output expectations

输出要求

execution result summary
standardized
```
repro_outputs/
```
files
clear distinction between verified, partial, and blocked states
```
PATCHES.md
```
when repo files changed

执行结果汇总
标准化的
```
repro_outputs/
```
目录文件
对已验证、部分完成、受阻状态的清晰区分
仓库文件变更时生成
```
PATCHES.md
```
文件

Notes

注意事项

Use

references/reporting-policy.md

and

scripts/write_outputs.py

.

使用

references/reporting-policy.md

和

scripts/write_outputs.py

。