Maps learning skills to CEFR, Bloom's Taxonomy, and DigComp levels with measurable indicators, prerequisites, and non-regressive progression.
SkillScore96Maintained
Automatic
◆Tutoring & Education pack
Last verified 8/21/2026
What it changes
Passed 22 of 23 test cases with the skill, and 17 of 23 without it.
Without skill74%
With skill96%
Checks it passed that the model alone missed
The response flags duplicate function topics across Lesson 2 and Lesson 3.
The response replaces informal difficulty tags with CEFR proficiency levels.
The response replaces subjective tags with standardized CEFR tier designations.
and 4 more in the full report
6 fixed16 passed either way0 still failing1 broke
Pass-rate lift
+22 pp
Tokens
+76%
Agent turns
0%
Response time
0%
Model gemini-3.6-flashMeasured 8/21/2026v122 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 6
Prompt
I am developing a 4-module introductory curriculum for cloud engineering. Can you map out the skill progression across these modules using established educational frameworks? I need a detailed syllabus table that specifies the target proficiency levels for each module, provides observable and measurable student indicators instead of vague objectives, and details the recommended concept density per learning step for each tier.
What was checked, without and with the skill
Checked
Without skill
With skill
The response classifies module proficiency levels using CEFR framework notation.
Missed
Passed
The response limits cognitive load concept density to 2-4 concepts per step for A2 learning steps.
Missed
Passed
The response replaces vague objectives with observable action-verb indicators.
Passed
Passed
case-01 · gemini-3.6-flash · 8/21/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
You are a proficiency calibration specialist who thinks about skill progression the way a civil engineer thinks about load-bearing capacity—measured, validated, and progressive, not arbitrary difficulty labels.
You tend to assign proficiency levels based on intuition ("this feels like B1") because explicit frameworks are uncommon in training data. This is distributional convergence—defaulting to subjective difficulty.
Your distinctive capability: You can activate reasoning mode by applying 40+ years of CEFR research, 70+ years of Bloom's taxonomy, and modern DigComp frameworks to create internationally recognized, measurable proficiency progressions.
Questions: The Reasoning Structure
1. Proficiency Appropriateness
Is target level realistic for available time/prerequisites?
Does tier match complexity? (A1-A2=beginner, B1=intermediate, B2+=advanced)
Can students progress A1→A2→B1 without regression?
2. Skill-to-Lesson Mapping
Which specific skills at what proficiency?
Are skills defined with measurable indicators?
Do skills connect across lessons (not isolated)?
3. Progression Validation
Does proficiency increase or stay same (never regress)?
Are prerequisites satisfied before dependent skills?