[workshop-sim] Repair: Fix checkpoint-quality detection regex to match :white_check_mark: shortcode

[github-actions]

Problem statement

The shared curriculum quantitative assessment rubric (.github/skills/curriculum-quantitative-assessment/curriculum_assessment.py) scores checkpoint_quality as 0.0/10 on all 30 main workshop steps, even though every main step ends with a ## :white_check_mark: Checkpoint section containing a well-formed checklist (verified directly, e.g. 07-your-first-workflow.md:115-122, 04-github-actions-intro.md:138, 08-run-your-workflow.md:99). The root cause is CHECKPOINT_RE = re.compile(r"##\s+✅\s*Checkpoint", re.IGNORECASE), which only matches a literal ✅ character, while every main step uses the GitHub emoji shortcode :white_check_mark: instead. This suppresses the entire checkpoint_quality dimension corpus-wide and drags the cohort-wide Learning KPI index down to 2.88/10 even though genuine checkpoints are present and well-formed.

Proposed change

Update CHECKPOINT_RE in curriculum_assessment.py to match both the literal emoji and the :white_check_mark: shortcode, e.g. re.compile(r"##\s+(?:✅|:white_check_mark:)\s*Checkpoint", re.IGNORECASE). Re-run the assessment across the corpus after the fix to confirm checkpoint_quality scores now reflect the checklist_items count already computed per step (currently unused because has_checkpoint is always False).

Failure mode classification

Learning barrier (measurement/detection gap masking real scaffolding quality, not a content gap) — this is a rubric bug fix, not a workshop content change.

Quantitative guardrail

• Current overall_score (corpus mean): 6.38/10; this fix is expected to raise, not lower, every step's overall_score since checkpoint_quality is weighted 2.0 in the rubric and currently floors at 0.0 everywhere
• Weakest rubric dimension corpus-wide: checkpoint_quality (0.0/10 on all 30 steps)
• Current learning KPI index (cohort mean): active_learning 4.18, checkpoint_quality 0.0, scaffolding 5.0 → index = 2.88/10
• Because this only corrects detection (no workshop content changes), no step's active_learning or scaffolding score is affected, and overall_score/learning KPI index can only increase or stay flat for steps with checklist items, never decrease

Acceptance criteria

• The shared curriculum quantitative assessment overall_score improves (or stays flat) for every step with an existing checklist, verified by re-running the assessment script before and after the regex fix
• The cohort-wide learning KPI index (2.0 × active_learning + 2.0 × checkpoint_quality + 1.5 × scaffolding) / 5.5 increases from its current 2.88/10 baseline
• No existing workshop checkpoint sections, checklists, or scaffolding content are modified — only the detection regex in curriculum_assessment.py changes
• Existing unit tests for curriculum_assessment.py (if any) pass, and a regression test is added confirming both ✅ and :white_check_mark: headings are detected

Suggested owner

copilot coding agent
Related to #3139

Generated by 🔬 Workshop Student Simulator · copilot · auto · 231.3 AIC · ⌖ 7.23 AIC · ⊞ 15.2K · ◷

• expires on Sep 10, 2026, 10:50 PM UTC

[github-actions]

This issue is being closed as outdated. A newer issue has been created: #3148

View newer issue

---

This action was performed automatically by the Workshop Student Simulator workflow.

[github-actions]

This issue is being closed as outdated. A newer issue has been created: #3149

View newer issue

---

This action was performed automatically by the Workshop Student Simulator workflow.