[curriculum-eval] side-quest-11-06-anthropic-key.md: active_learning — procedural steps as passive numbered lists instead of checkable actions

[github-actions]

File: workshop/side-quest-11-06-anthropic-key.md
Overall Score: 5.39 / 10.0 (corpus mean: 6.15)

Flagged Dimensions:

Dimension	Score	Benchmark	Delta
cognitive_load	7.2	≤ 15 new concepts	new_concepts 26 (-5.5 pts, ~1.7× budget)
active_learning	3.5	≥ 3 density	-6.5

Root Cause (≤ 2 sentences):
At only 665 words (well under the 800-word budget) the page still introduces 26 new concepts — API key generation, repository secrets UI navigation, secret naming rules, frontmatter fields (engine: claude, network.allowed, copilot-requests: write removal), and gh aw compile — because it interleaves procedural steps (click paths through the Anthropic console and GitHub Settings UI) with conceptual frontmatter knowledge, each treated as a distinct new term for a learner following this side quest for the first time. The 2 code blocks against 5 checklist items yield activity density of only 1.05, well below the ≥3 benchmark, despite the page being almost entirely a step-by-step procedure.

Evidence (quoted from the file):

1. Go to console.anthropic.com and sign in... 2. Click Create Key... 3. Copy the key value...

(followed by) 1. Open your repository on GitHub. 2. Click Settings → Secrets and variables → Actions. 3. Click New repository secret...

Learning Science Rationale:
Although the numbered steps look like "active learning" to a human reader, the rubric's activity_density metric only counts code blocks and checklist items per 100 words — this page's procedure is delivered as plain numbered lists (UI click-paths) rather than as inline commands or verifiable checklist items the learner ticks off as they go, so Mayer's principle of interactivity (learners should actively confirm each step, not just read it) is under-served: there is no way for the learner to self-check "did I do this correctly" until the final checkpoint, five distinct procedures later.

Improvement Prompt (for an agent):

Edit workshop/side-quest-11-06-anthropic-key.md to raise activity density and reduce concept-switching cost:
1. Convert the "Get an Anthropic API key" and "Store the key as a repository secret" numbered lists into inline checklist items (- [ ] ...) matching the style of the final Checkpoint section, so learners can tick off each sub-step as they complete it rather than reading a passive numbered list.
2. After the "Update your workflow frontmatter" code block, add a short inline verification command or check (e.g. a `grep` or visual diff callout) so the learner confirms the frontmatter change took effect before running `gh aw compile`, giving an intermediate self-check rather than jumping straight to compile.
3. Do not add new concepts; only restructure existing procedural content into checklist form and add the one verification step described above.
4. Run `npx --yes markdownlint-cli2 "workshop/side-quest-11-06-anthropic-key.md"` after editing and fix any lint errors.

Expected Score After Fix: 6.3 / 10.0

Generated by 🔬 Curriculum Quality Evaluator · copilot · auto · 62.1 AIC · ⌖ 23.2 AIC · ⊞ 9.5K · ◷

• expires on Sep 10, 2026, 3:45 PM UTC