[curriculum-eval] side-quest-01-02-environment-reference.md: active_learning — passive glossary page with almost no retrieval practice

**File:** `workshop/side-quest-01-02-environment-reference.md`
**Overall Score:** `4.93 / 10.0` (corpus mean: `6.14`)

**Flagged Dimensions:**

| Dimension | Score | Benchmark | Delta |
|---|---|---|---|
| checkpoint_quality | 0.0 | 10 (≥4 checklist items) | -10.0 |
| active_learning | 2.1 | ≥3 activity density | -0.9 (density 0.63 vs 3.0 ideal) |
| cognitive_load | 6.3 | 10 (≤800 words, ≤15 concepts) | -3.7 (1433 words, 17 concepts) |

**Root Cause (≤ 2 sentences):**
This reference page is a 1433-word glossary of nine tools/environments, each illustrated with a screenshot and a short caption but almost no hands-on interaction (only 2 code blocks across the whole page), which drives activity density to 0.63 against the 3.0 target. The rubric also scores `checkpoint_quality` as 0.0 because its detector only matches the literal `## ✅ Checkpoint` heading, while this file (like 87 others) correctly uses the repo's `## :white_check_mark: Checkpoint` shortcode convention — this is a scorer defect, not a content defect, and should not drive the content fix.

**Evidence (quoted from the file):**
> Knowing which name maps to which role helps you follow workshop instructions without stopping to wonder what "the terminal" or "Codespaces" means in context.
>
> These visuals are simplified mental models, not literal product screenshots. Use them to recognize what each name refers to when it appears in later steps.

**Learning Science Rationale:**
Mayer's Multimedia Principles hold that pairing static images with passive prose (the "picture + caption" pattern repeated nine times here) produces recognition, not retention, unless learners actively process the material — e.g., by predicting, matching, or self-testing before seeing the answer. Right now the page asks learners to *read* nine tool descriptions in sequence with no retrieval practice between them, so despite a large amount of content there is very little active learning per Sweller's Cognitive Load Theory, which predicts that passive exposure without spaced retrieval yields weak long-term encoding relative to the same time spent on interactive review.

**Improvement Prompt (for an agent):**
```
Open workshop/side-quest-01-02-environment-reference.md. Convert the "Environment and tool glossary" and "Conceptual screenshots" sections from passive read-only prose into an active-recall exercise: after the glossary table, add a short "Quick self-check" activity (a ``-wrapped matching or fill-in-the-blank exercise, e.g. "Match each tool name to its role before revealing the answer") so learners retrieve the term-to-role mapping instead of only rereading it. Keep the existing screenshots but reduce the per-tool caption prose so overall word count drops toward 1000-1100 words. Do not remove the existing Checkpoint section. Run `npx --yes markdownlint-cli2 "**/*.md"` after editing and fix any lint errors.
```

**Expected Score After Fix:** `6.5 / 10.0`

> Generated by [🔬 Curriculum Quality Evaluator](https://github.com/githubnext/gh-aw-workshop/actions/runs/35026063971) · copilot · auto · 76.1 AIC · ⌖ 8.13 AIC · ⊞ 9.5K · [◷](https://github.com/search?q=repo%3Agithubnext%2Fgh-aw-workshop+is%3Aissue+%22gh-aw-workflow-call-id%3A+githubnext%2Fgh-aw-workshop%2Fcurriculum-evaluator%22&type=issues)
> - [x] expires on Sep 16, 2026, 9:39 PM UTC

Open workshop/side-quest-01-02-environment-reference.md and review the “Environment and tool glossary” and “Conceptual screenshots” sections first. Add the requested details-wrapped matching or fill-in-the-blank self-check, reduce caption prose toward 1000–1100 words, preserve the existing Checkpoint section, and run npx --yes markdownlint-cli2 "**/*.md" until lint passes.

Written by the indexing model from the issue text.