Learning Objectives Skill

Evals-First Objective Design (Constitution v4.0.1)

CRITICAL WORKFLOW:

1. Evals First: Review success criteria from chapter spec BEFORE writing objectives
2. Objectives Second: Design learning objectives that lead to eval success
3. Alignment Third: Ensure each objective maps to at least one success eval
4. Validation Fourth: Verify objectives are measurable and aligned

Template:

### Objective Design (Evals-First)

**Source**: Chapter spec at `specs/part-X/chapter-Y/spec.md`

**Success Evals from Spec**:
1. 75%+ write valid specification (business goal: reduce iteration cycles)
2. 80%+ identify vague requirements (business goal: prevent scope creep)

**Learning Objectives Designed to Achieve Evals**:
- LO-001: Write clear specifications → Eval #1
- LO-002: Identify ambiguous requirements → Eval #2

Do NOT create objectives without:

• ✅ Reference to approved spec with success evals
• ✅ Explicit mapping: Objective → Eval → Business Goal
• ✅ Measurability aligned with eval criteria

Process

Step 1: Understand the Context

When a request comes in to generate learning objectives, first understand:

• What topic or concept? (Python decorators, OOP, async/await, etc.)
• Who are the learners? (Beginners, intermediate, advanced)
• How long to teach? (30 minutes, 2 hours, full unit)
• What's the end goal? (Can they build something? Understand theory? Debug code?)
• What are the success evals? (From chapter spec—what defines success?)

Step 2: Review Bloom's Taxonomy (If Needed)

If you're not familiar with the specific topic's cognitive levels, read: 📖 reference/blooms-taxonomy-programming.md

This document maps Bloom's 6 levels to programming contexts with:

• Action verbs for each level (Remember, Understand, Apply, Analyze, Evaluate, Create)
• Programming examples
• Assessment methods for each level

Step 3: Identify Prerequisites

Read the guidance on prerequisite analysis: 📖 reference/prerequisite-analysis.md

For your objectives, determine:

• What must learners know BEFORE tackling the main concept?
• List prerequisites at Remember/Understand level (not deep mastery)
• Trace dependency chains to foundational knowledge

Step 4: Choose Assessment Methods

Based on the Bloom's level of each objective, review appropriate assessment methods: 📖 reference/assessment-methods.md

This guides you to pair objectives with realistic assessment approaches (code exercises for Apply level, code reviews for Evaluate, etc.).

Step 5: Generate Objectives with CEFR Proficiency Levels

For each topic, create 3-5 objectives (typically):

• At least one from each level needed for the topic (Remember through Create)
• Progressively building in complexity
• Each with clear statement, context, prerequisites, assessment method, and success criteria
• Map to CEFR proficiency level (A1/A2/B1/B2/C1)

Use the template as guidance: 📄 templates/learning-objective-template.yml

Key principle: Each objective should answer:

• What will learners DO? (verb from Bloom's level)
• In what context? (the specific situation or problem)
• How will we know they succeeded? (measurable criteria)
• What proficiency level? (CEFR A1-C2)

CEFR Proficiency Mapping (Constitution v3.1.2)

Align objectives with international proficiency standards (from skills-proficiency-mapper v2.0):

A1 (Beginner - Recognition):

• Bloom's: Remember/Understand only
• Example: "Identify Python syntax for defining a function"
• Measurable: Recognition, not production

A2 (Elementary - Guided Application):

• Bloom's: Understand/Apply with scaffolding
• Example: "Complete a function definition with provided hints"
• Measurable: Application with support

B1 (Intermediate - Independent Application):

• Bloom's: Apply independently
• Example: "Implement a function from clear specification without assistance"
• Measurable: Real-world application without scaffolding

B2 (Upper-Intermediate - Analysis):

• Bloom's: Analyze/Evaluate
• Example: "Compare two implementations and justify which is more maintainable"
• Measurable: Evaluation with justification

C1 (Advanced - Creation/Synthesis):

• Bloom's: Evaluate/Create
• Example: "Design a system architecture for scalable deployment"
• Measurable: Original design with trade-off analysis

Proficiency Progression Rule: Lessons should progress A1→A2→B1 within a chapter (not jump from A1 to C1).

Three-Role AI Partnership Objectives (Section IIb, Constitution v4.0.1)

CRITICAL: AI-native learning objectives must include ability to work WITH AI in bidirectional co-learning partnership (per Section IIb forcing functions), not just independently.

Traditional Objective Format:

LO-001: Implement user authentication (independent skill)

AI-Native Objective Format:

LO-001: Implement user authentication working with AI as co-learning partner
  - Use AI as Teacher: Learn security patterns from AI suggestions
  - Use AI as Student: Refine AI's output through clear specifications
  - Use AI as Co-Worker: Iterate toward optimal solution collaboratively
  - Validate: Verify AI-generated code meets security requirements

Three-Role Objective Types:

1. AI as Teacher Objectives (Student learns from AI):

• "Identify pattern suggested by AI that improves code quality"
• "Explain trade-offs in AI's proposed approaches"
• "Apply AI-suggested pattern to new context"

2. AI as Student Objectives (Student teaches AI):

• "Write specification that produces correct code on first try"
• "Provide feedback that improves AI's next iteration"
• "Clarify requirements when AI asks for disambiguation"

3. AI as Co-Worker Objectives (Collaborative iteration):

• "Iterate with AI to converge on optimal solution"
• "Make strategic decisions while AI handles tactical implementation"
• "Validate AI outputs for correctness and appropriateness"

Example AI-Native Objective Set:

- id: "LO-AUTH-001"
  statement: "Implement OAuth authentication working with AI as co-learning partner"
  blooms_level: "Apply"
  cefr_level: "B1"
  three_role_integration:
    ai_as_teacher: "Learn refresh token rotation pattern from AI suggestion"
    ai_as_student: "Guide AI through security requirements via clear spec"
    ai_as_coworker: "Iterate on session management approach together"
  assessment_method: "Code + reflection: Show implementation AND what you learned from AI"
  success_criteria:
    - "OAuth implementation works correctly"
    - "Student identifies at least one pattern learned from AI"
    - "Student demonstrates validation of AI output"

Objective Balance for AI-Native Content:

• 60-70%: Traditional technical skills
• 20-30%: Co-learning skills (working WITH AI)
• 10-20%: Validation/verification skills

Step 6: Validate for Measurability

Once you've generated objectives, invoke the validation script to check they're measurable:

python .claude/skills/learning-objectives/scripts/validate-objectives.py objectives.yml

The script checks:

• ✅ Objectives use action verbs (not vague words like "understand" or "know")
• ✅ Objectives are specific and testable
• ✅ Verbs match the Bloom's level
• ✅ Required fields present (context, assessment, success criteria)

If validation fails, review the error messages and refine:

• Replace vague verbs with specific action verbs from Bloom's level
• Add missing context or success criteria
• Make statements more specific (not just topic names)

Step 7: Refine and Finalize

Based on validation feedback:

• Revise any objectives with errors
• Ensure the objective sequence makes sense (build progressively)
• Verify prerequisites are necessary but not excessive
• Check that assessment methods are realistic for the time available

Output Format

Present learning objectives in clear, structured format: