lab-scaffolding: Скилл для создания учебных лабораторий

--- name: lab-scaffolding description: "A lab scaffolding skill used by the lab-designer agent. Provides lab difficulty calibration, starter code design, capstone project structures, and self-directed learning strategies. Used for 'lab design,' 'project assignments,' 'scaffolding,' 'hands-on labs,' and related topics." --- # Lab Scaffolding — Lab Scaffolding Methodology Expert knowledge used by the lab-designer agent when designing hands-on assignments and projects. ## Why Scaffolding If you tell learners "build a program" without support, 90% will give up. **Scaffolding** is the strategy of gradually providing support for what learners cannot do alone, then reducing support as their ability grows. ## Five-Level Lab Pyramid ``` / / 5 <- Capstone Project (open-ended) /----- / 4 <- Mini Project (minimal guidance) /----------- / 3 <- Challenge (hints provided) /----------------- / 2 <- Guided Lab (step-by-step) /--------------------- / 1 <- Follow-Along (code walkthrough) /--------------------------- ``` ### Design by Level | Level | Learner Autonomy | Support Level | Lab Format | |-------|-----------------|---------------|-----------| | **1. Follow-Along** | 0% | 100% guided | Type instructor's code verbatim | | **2. Guided Lab** | 20% | 80% guided | Starter code + fill in the blanks | | **3. Challenge** | 50% | Hints only | Requirements -> Hints -> Verification | | **4. Mini Project** | 80% | Spec only | Feature requirements -> free implementation | | **5. Capstone** | 95% | Mentoring only | Choose topic -> Design -> Implement -> Present | ## Starter Code Design ### Starter Code Principles 1. **Runnable state**: Starter code must execute without errors as-is 2. **TODO markers**: Clearly mark where learners should write code 3. **Remove boilerplate**: Pre-write code unrelated to learning objectives 4. **Progressive disclosure**: Provide more code early on, bare files later ### Starter Code Template ```python """ Lab: [Lab Title] Learning Objective: [Skills to practice in this lab] Estimated Duration: XX minutes Difficulty: [1-5] """ # === Pre-written code (do not modify) === import flask app = flask.Flask(__name__) # === TODO 1: [Specific instruction] === # Hint: [Related concept or method to use] # Expected lines of code: X # === TODO 2: [Specific instruction] === # Hint: [Related concept or method to use] # === Tests === # If the code below runs without errors, you've succeeded if __name__ == "__main__": # Automated verification code pass ``` ## Capstone Project Design ### Capstone Structure Template ``` ## Capstone Project: [Title] ### Overview [Purpose of the project and description of final deliverable] ### Learning Objective Mapping This project comprehensively assesses the following learning objectives: - [LO1]: [Learning objective] - [LO2]: [Learning objective] - ... ### Functional Requirements #### Must Have 1. [Feature 1] 2. [Feature 2] 3. [Feature 3] #### Nice to Have — Bonus Points 1. [Additional feature 1] 2. [Additional feature 2] ### Milestones | Week | Deliverable | Checkpoint | |------|------------|-----------| | Week 1 | Design document + wireframes | Mentor review | | Week 2 | Core feature prototype | Working demo | | Week 3 | Full feature implementation | Code review | | Week 4 | Testing + documentation + presentation | Final submission | ### Rubric [Reference assessment-engineering skill rubric] ### Reference Resources - [Related lesson links] - [Official documentation links] - [Example projects (reference only, no copying)] ``` ### Capstone Topic Design Rules 1. **Real-world problems**: Have learners build something they can actually use 2. **Appropriate scope**: Completable in 2-4 weeks 3. **Learning objective integration**: Must use all core skills taught in the course 4. **Scalable**: Required features are achievable; bonus features adjust difficulty 5. **Portfolio value**: The finished product should be resume-worthy ## Difficulty Calibration Techniques ### Dynamic Difficulty Adjustment | Situation | Strategy | |-----------|---------| | Learner is stuck | Hint 1 (direction) -> Hint 2 (specific) -> Hint 3 (partial code) -> Full solution | | Too easy for learner | Add constraints, optimization challenges | | Wide skill gap among learners | Separate required/optional tasks; assign mentoring roles to advanced learners | ### Hint System Design ``` [Attempt without hints] -> After 5 min [Hint 1: Direction] "This problem uses the ~ concept" -> After 5 min [Hint 2: Approach] "Look up the ~ method" -> After 5 min [Hint 3: Code skeleton] "The structure looks like this: ..." -> After 5 min [Full solution + explanation] ``` ## Recommended Lab Environments | Course Type | Recommended Environment | Reason | |------------|------------------------|--------| | Programming Intro | Browser IDE (Replit, Google Colab) | Start instantly without installation | | Web Development | GitHub Codespaces / Gitpod | Consistent environment guaranteed | | Data Analysis | Google Colab / Kaggle Notebooks | GPU access, dataset hosting | | Systems/Infrastructure | Docker + scripts | Reproducible environment |