File: workshop/05-agentic-workflows-intro.md
Overall Score: 5.43 / 10.0 (corpus mean: 6.14)
Flagged Dimensions:
| Dimension |
Score |
Benchmark |
Delta |
| cognitive_load |
8.5 |
≤ 800 words |
word_count=1107 (−3.0 vs ideal) |
| active_learning |
2.4 |
density ≥ 3 |
activity_density=0.72 (2 code blocks + 6 checklist items / 11.07×100 words) — lowest active_learning score among all flagged files |
| checkpoint_quality |
0.0 |
— |
tooling false negative (see companion regex-fix issue); page has a valid checkpoint with 6 items |
Root Cause (≤ 2 sentences):
Despite containing three well-designed "Activity" sections (predict-then-check diagram, agentic-vs-standard classification, write-your-own task brief), the page's 1107-word length and low ratio of code blocks to prose (only 2 fenced blocks across 3 activities, one of which is a comment-only placeholder) keeps activity_density well below the ≥3 benchmark, meaning the interactive elements aren't registering as "activity" density despite genuine pedagogical engagement design.
Evidence (quoted from the file):
Activity 1 — identify the parts: Open any .lock.yml file in your repo and find the on: key. That is the compiled trigger that came from your frontmatter.
# Example: open .github/workflows/my-workflow.lock.yml
# Find the "on:" key — that is your compiled trigger.
Learning Science Rationale:
Mayer's Multimedia Principles favor concise, segmented content paired with hands-on retrieval tasks; here the ratio of explanatory prose (three terms table, two-file-structure diagram, lifecycle diagram) to countable "activity" artifacts (code fences, checklist items) is skewed toward passive reading, so even though the page has three good Activities in prose form, the rubric's density formula—driven by code_blocks and checklist_items—undercounts prose-based prediction exercises as active learning.
Improvement Prompt (for an agent):
Edit workshop/05-agentic-workflows-intro.md to raise measurable activity density and trim length:
1. Convert "Activity 2 — agentic or standard?" (Task A / Task B) into a short numbered or checklist-style quiz format using `- [ ]` items for each classification decision before revealing the answer, so the exercise is countable as active learning content, e.g.:
- [ ] Task A (lint/tests on every PR) is: standard / agentic
- [ ] Task B (daily issue triage summary) is: standard / agentic
2. Convert "Activity 3 — write a task brief" placeholder code block into a checklist of the three evaluation criteria (time window, output format, priority signal) as `- [ ]` items instead of prose bullets, so learners can mark each criterion as they check their own brief.
3. Trim the "Three key terms" and "two-file structure" sections by removing one of the two diagrams if both convey overlapping information (compare the three-terms diagram and the lifecycle diagram) — keep only the lifecycle diagram since it is referenced by the compile activity that follows.
Keep the existing checkpoint section unchanged.
Expected Score After Fix: 6.2 / 10.0
Generated by 🔬 Curriculum Quality Evaluator · copilot · auto · 77.1 AIC · ⌖ 14.6 AIC · ⊞ 9K · ◷
File:
workshop/05-agentic-workflows-intro.mdOverall Score:
5.43 / 10.0(corpus mean:6.14)Flagged Dimensions:
Root Cause (≤ 2 sentences):
Despite containing three well-designed "Activity" sections (predict-then-check diagram, agentic-vs-standard classification, write-your-own task brief), the page's 1107-word length and low ratio of code blocks to prose (only 2 fenced blocks across 3 activities, one of which is a comment-only placeholder) keeps activity_density well below the ≥3 benchmark, meaning the interactive elements aren't registering as "activity" density despite genuine pedagogical engagement design.
Evidence (quoted from the file):
Learning Science Rationale:
Mayer's Multimedia Principles favor concise, segmented content paired with hands-on retrieval tasks; here the ratio of explanatory prose (three terms table, two-file-structure diagram, lifecycle diagram) to countable "activity" artifacts (code fences, checklist items) is skewed toward passive reading, so even though the page has three good Activities in prose form, the rubric's density formula—driven by code_blocks and checklist_items—undercounts prose-based prediction exercises as active learning.
Improvement Prompt (for an agent):
Expected Score After Fix:
6.2 / 10.0