Model A/B replay on a finished task

未关闭
#175 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
5/5
预计耗时
一周以上
新手友好度
35/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
活跃
技术栈
typescript
领域
devtools

调研方向

Start by reading the implementation and tests for task replay, provider selection, sandbox creation, and Hindsight, along with related issues #28 and #145 and spec E25. The issue does not name files or entry points, so first map where finished tasks, base commits, and provider runs are handled. Done means replay starts from the recorded base commit without workspace side effects and the side-by-side comparison shows diff, tests, tokens, and time.

由索引模型根据 Issue 内容生成。

描述

area:providers area:sandbox needs-triage p2 type:feature

Part of #145 · beat-Paperclip spec id E25

  • Priority: p2
  • Labels: type:feature, area:providers, area:sandbox
  • Relates to: #28, #145
  • Scope: "Re-run this task with…" any discovered provider, in a sandbox, from the same starting commit. The two runs are compared side by side, and the verdict is retained in Hindsight to inform model policies (E13).
  • Acceptance:
    • A replay starts from the recorded base commit, with no workspace side effects.
    • The comparison shows diff, tests, tokens and time.

Source: Paperclip issue/discussion research (docs, GitHub issues, live instance).

主要语言
TypeScript
星标
3
派生
0
平均合并
17 小时 33 分钟
30 天内合并 PR
163

环境准备

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

musterhq/muster-code 的其他 Issue

查看 musterhq/muster-code 的全部 Issue

相似的 Issue

更多 TypeScript Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。