review

Context

• Learning directory: !ls -d .learning 2>/dev/null
• Current topic: !ls .learning/ 2>/dev/null

Note: If .learning/ doesn't exist, inform user to run /learn [topic]. Check topic folders (ignore scripts/).

Your Task

Conduct spaced repetition reviews to combat forgetting and reinforce learning.

If no .learning/:

• Inform: "No learning in progress. Use /learn [topic] to start!"

If reviews due:

For each concept in review list:

1. Present: "Let's review: [Concept Name]"
2. Prompt teaching (rotate):
   • "Explain [concept] in your own words"
   • "How would you teach [concept] to a beginner?"
   • "What's the key idea behind [concept]?"
3. Listen to user's explanation
4. Evaluate:
   • Clear & accurate → Praise, mark reviewed
   • Partial → Ask clarifying questions, guide to fill gaps
   • Incorrect → Gently correct, provide hints
5. Mark reviewed: python3 .learning/scripts/review_scheduler.py review <topic-slug> "[Concept]"

After all reviews:

• Celebrate: "Great job! Reviewed N concepts! 🎉"
• Show next review date
• Use AskUserQuestion for next action:

{
  "question": "What would you like to do next?",
  "header": "Next",
  "multiSelect": false,
  "options": [
    {
      "label": "Learn new",
      "description": "Continue with next syllabus item"
    },
    {
      "label": "Practice",
      "description": "Work on hands-on exercises"
    },
    {
      "label": "Take break",
      "description": "Come back later"
    }
  ]
}

If no reviews due:

• Inform: "No reviews due today! Next: [date]"
• Suggest continuing with new material

Handling forgotten concepts:

• Don't give answer immediately
• Provide hints: "It's related to [context]..."
• If still stuck: Review briefly, reschedule for tomorrow
• Reschedule: python3 .learning/scripts/review_scheduler.py add <topic-slug> "[Concept]"

Key principle: Active recall (user reconstructs from memory), not passive recognition