[workshop-sim] Repair: Add active-recall checkpoint to break up concept density in 05-agentic-workflows-intro.md

[github-actions]

Problem statement

05-agentic-workflows-intro.md has the highest conditional dropout rate in the entire curriculum (20.7% among 36,906 at-risk simulated runs, 95% MC interval 20.3%–21.2%) and the highest cognitive_load score (8.5/10) of any page. The failure categories driving this dropout are agentic-concept-gap (4,644 failures) and deployment-capability-gap (2,196 failures) — learners can absorb the vocabulary but cannot yet connect it to what they will actually build and deploy. The page introduces three abstract terms (trigger, task brief, safe outputs), a two-file lifecycle model, and two multi-part decision activities back-to-back with no intermediate recall checkpoint between the terms table and the two-file-structure section.

Proposed change

Insert one short, focused active-recall check immediately after the "Three key terms" table and before the "The two-file structure" section — a 2-3 question self-check (e.g., "Which of the three terms controls how the workflow writes back to GitHub?") with a <details>-hidden answer. This breaks up the concept load into smaller digestible chunks without removing any existing content, activities, or the two decision exercises already on the page.

Failure mode classification

Learning barrier (conceptual density / insufficient scaffolding between concept introductions — not an access/tooling barrier).

Quantitative guardrail

Current scores for 05-agentic-workflows-intro.md:

• overall_score: 5.43/10
• Weakest rubric dimensions: checkpoint_quality 0.0/10 (see note below — likely a rubric detection artifact, not a true scaffolding gap) and active_learning 2.4/10
• Learning KPI index: (2.0×2.4 + 2.0×0.0 + 1.5×5.0)/5.5 = 2.24/10

Adding a short recall checkpoint directly targets active_learning (more low-friction practice opportunities) and scaffolding (breaking one dense concept block into two smaller ones), which should raise both dimensions and the overall/KPI scores. Note: the checkpoint_quality metric currently scores 0.0 for every step in the corpus because the rubric's CHECKPOINT_RE only matches a literal ✅ character while the page uses the :white_check_mark: shortcode — this repair should not be blocked on that separate rubric-tooling issue, but the new recall check should still use the page's existing heading conventions.

Acceptance criteria

• The shared curriculum quantitative assessment overall_score for 05-agentic-workflows-intro.md stays flat or improves after the change (baseline: 5.43/10)
• The step's learning KPI index (2.0 × active_learning + 2.0 × checkpoint_quality + 1.5 × scaffolding) / 5.5 stays flat or improves (baseline: 2.24/10)
• All existing hands-on activities (the two decision exercises, task-brief-writing exercise, and end-of-page checklist) are preserved unchanged — the new recall check is additive, not a replacement
• The new recall check is placed between the "Three key terms" table and "The two-file structure" heading, uses the same <details>/reveal pattern already used elsewhere on the page, and does not add more than 2-3 questions

Suggested owner

copilot coding agent
Related to #3152

Generated by 🔬 Workshop Student Simulator · copilot · auto · 271 AIC · ⌖ 27.2 AIC · ⊞ 15.2K · ◷

• expires on Sep 11, 2026, 3:18 AM UTC