Skip to content

Commit 9221938

Browse files
committed
Polish README copy nuances
1 parent e0d518f commit 9221938

2 files changed

Lines changed: 6 additions & 6 deletions

File tree

README.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
</p>
88

99
<p align="center"><strong>Turn AI coding runs into portable, replayable task bundles.</strong></p>
10-
<p align="center">Useful when chat logs are too loose and full benchmark platforms are too heavy.</p>
10+
<p align="center">Useful when chat logs are too loose and full benchmark platforms are more than you need.</p>
1111
<p align="center">
1212
<a href="#quickstart"><strong>Quick Start</strong></a> ·
1313
<a href="#example-output"><strong>Example Output</strong></a> ·
@@ -29,7 +29,7 @@ Use it to:
2929
- generate benchmark-style reports from a folder of bundles
3030
- preserve enough context for reruns without aiming for token-perfect replay
3131

32-
It works best when you want to:
32+
Reach for it when you want to:
3333
- inspect what happened
3434
- share a task with someone else
3535
- rerun a task later
@@ -125,7 +125,7 @@ See the committed sample report:
125125

126126
Most AI coding work disappears into screenshots, transcripts, or one-off patches.
127127

128-
Task Bundle gives you a durable unit you can inspect, archive, compare, validate, and report on. That makes it useful for:
128+
Task Bundle gives you a stable task artifact you can inspect, archive, compare, validate, and report on. That makes it useful for:
129129
- agent builders who want reproducible tasks
130130
- eval and benchmark authors who need structured task artifacts
131131
- teams comparing Codex, Claude Code, Cursor, or custom tools

README.zh-CN.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
</p>
88

99
<p align="center"><strong>把 AI coding 过程整理成可分享、可比较、可重跑的任务包。</strong></p>
10-
<p align="center">适合用在聊天记录不够稳定、完整 benchmark 平台又太重的场景里。</p>
10+
<p align="center">适合用在聊天记录太散、完整 benchmark 平台又太重的时候。</p>
1111
<p align="center">
1212
<a href="#quickstart"><strong>快速开始</strong></a> ·
1313
<a href="#example-output"><strong>示例输出</strong></a> ·
@@ -27,7 +27,7 @@ Task Bundle 是一个 TypeScript + Node.js CLI,用来把一次编码任务打
2727
- 把任务输入、执行摘要、diff、事件和工作区文件放在一起
2828
- 比较 Codex、Claude Code、Cursor 或内部工具的运行结果,并保留元数据、哈希和 outcome 字段
2929
- 从一组 bundle 生成 benchmark 风格报告
30-
- 为后续重跑保留足够上下文,而不是追求逐 token 回放
30+
- 为后续重跑保留足够上下文,而不是追求逐 token 一致性
3131

3232
它比较适合这些场景:
3333
- 查看一次任务最后到底做了什么
@@ -125,7 +125,7 @@ Ranking
125125

126126
很多 AI coding 结果最后只留下截图、聊天记录或者一个 patch,后续几乎没法稳定比较。
127127

128-
Task Bundle 主要解决的是这个问题:把一次任务变成一个可以查看、归档、比较、校验、生成报告的稳定单元。它比较适合:
128+
Task Bundle 想解决的就是这个问题:把一次任务变成一个可以查看、归档、比较、校验、生成报告的稳定单元。它比较适合:
129129
- 想做可复现实验的 agent 作者
130130
- 想做任务评测和 benchmark 的团队
131131
- 想比较 Codex、Claude Code、Cursor 或内部工具的开发者

0 commit comments

Comments
 (0)