[curriculum-eval] curriculum_assessment.py: checkpoint_quality — regex misses :white_check_mark: headings, 92 pages falsely scored 0

未关闭 适合新手
  • 哪里不对？
#5,367 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 3 天内回复

还没有人认领这个 Issue。

评估

难度
2/5
预计耗时
1-3 小时
新手友好度
77/100
Issue 类型
缺陷
描述清晰度
描述清楚
活跃度
活跃
技术栈
github-actions, python
领域
testing-qa

调研方向

从.github/skills/curriculum-quantitative-assessment/curriculum_assessment.py开始，重点检查CHECKPOINT_RE和用于检测先决条件的正则表达式。将这些模式与 issue 中引用的标题进行比较，并检查相关测试或 fixture。完成标准是检测器能够识别 shortcode 标题，且引用的 workshop 页面能正确报告其 checkpoint；issue 还要求重新运行对workshop/*.md的评估。

由索引模型根据 Issue 内容生成。

描述

curriculum documentation quality

File: .github/skills/curriculum-quantitative-assessment/curriculum_assessment.py (affects every workshop/*.md)
Overall Score: n/a — rubric defect (corpus mean: 6.06)

Flagged Dimensions:

Dimension Score Benchmark Delta
checkpoint_quality 0.0 on 92/94 learning pages ≥ 7.5 (≥ 3 items) −7.5 or more
scaffolding 5.0 on pages that have "Before You Start" (e.g. side-quest-01-02-environment-reference.md) 10 −5.0

Root Cause (≤ 2 sentences):
CHECKPOINT_RE = re.compile(r"##\s+✅\s*Checkpoint", re.IGNORECASE) only matches the literal emoji, but workshop pages use the :white_check_mark: shortcode per repo conventions. The scorer therefore reports has_checkpoint: false and missing_checkpoint for pages that do have checkpoints, depressing the corpus mean by up to ~2 points (weight 2.0).

Evidence (quoted from the file):

## :white_check_mark: Checkpoint — workshop/04-github-actions-intro.md line 138, which lists five checklist items yet is reported as has_checkpoint: false, checkpoint_quality: 0.0.

Learning Science Rationale:
Valid measurement is a precondition for acting on rubric output. Formative-assessment research (Black & Wiliam) stresses that checkpoints give learners retrieval practice and feedback; a metric that cannot see them sends authors to add redundant checkpoints, or to ignore the signal entirely. Fixing the detector restores the rubric's ability to discriminate genuinely weak checkpoints (few or vague items) from present ones, and also exposes that history deltas such as the −2.00 drops on side-quest-11-09-agent-session-phases.md and side-quest-12-01-iterate-agent-output.md are likely heading-format changes rather than real content regressions.

Improvement Prompt (for an agent):

Edit .github/skills/curriculum-quantitative-assessment/curriculum_assessment.py:
1. Change CHECKPOINT_RE to r"##\s+(?:✅|:white_check_mark:)\s*Checkpoint" (case-insensitive).
2. Verify the prerequisite/scaffolding regex also matches "## :clipboard: Before You Start" and "Prerequisites" with the :clipboard: shortcode; fix if not (side-quest-01-02-environment-reference.md has this heading but has_prereq_section is false).
3. Update any unit tests/fixtures and re-run the assessment over workshop/*.md; confirm has_checkpoint is true for workshop/04-github-actions-intro.md and that the number of pages flagged missing_checkpoint drops from 92 to only pages genuinely lacking one.
4. Regenerate score-history baselines if they are cached so trend deltas stay meaningful.

Expected Score After Fix: 7.5 / 10.0 (corpus mean)

Generated by 🔬 Curriculum Quality Evaluator · copilot · auto · ~43 AIC · ⌖ ~2 AIC · ⊞ 5K · ◷

  • expires on Nov 6, 2026, 11:04 AM UTC
主要语言
JavaScript
星标
58
派生
40
平均合并
4 天 4 小时
30 天内合并 PR
8

环境准备

在 Codespaces 中打开

在浏览器里用你自己的 GitHub 账号启动这个项目的开发容器。

  • 没有 Dockerfile 或 Docker Compose 文件
  • 没有 Pull Request 模板
  • 没有贡献指南

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。

githubnext/gh-aw-workshop 的其他 Issue

查看 githubnext/gh-aw-workshop 的全部 Issue

相似的 Issue

更多 JavaScript Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。