AgentSkillsCN

spaced-repetition-coach

精通间隔重复法在面试备考中的应用。讲解颜色编码体系(灰/橙/黄/绿)、复习计划、合格阈值,以及每日筛选算法。适用于讨论问题颜色、复习时机、记忆策略,或解释间隔重复系统的运作原理时使用。

SKILL.md
--- frontmatter
name: spaced-repetition-coach
description: Expert on spaced repetition for interview prep. Explains the color-coding system (gray/orange/yellow/green), review schedules, eligibility thresholds, and the daily selection algorithm. Use when discussing problem colors, review timing, retention strategies, or how the spaced repetition system works.

Spaced Repetition Coach

Expert guidance on using spaced repetition effectively for LeetCode interview preparation.

Color-Coding System

The app uses an Anki-style color system to track problem mastery:

Color Progression

  • Gray (New): Never attempted or just added
  • Orange (Learning): Attempted but need more practice
  • Yellow (Reviewing): Getting comfortable, need periodic review
  • Green (Mastered): Confident, can solve efficiently

Color Transitions

Moving Forward:

  • Gray → Orange: First successful attempt
  • Orange → Yellow: Second successful attempt (after waiting period)
  • Yellow → Green: Third successful attempt (after longer waiting period)

Moving Backward (Regression):

  • If you fail to solve a problem correctly, move it back one color level
  • This ensures you focus on problems you're actually struggling with

Eligibility Thresholds

Problems become eligible for review based on color and time since last attempt:

ColorThresholdReasoning
GrayAlways eligibleNew problems always available
Orange3+ daysShort interval for recent learning
Yellow7+ daysMedium interval for consolidation
Green14+ daysLong interval for maintenance

Key Principle: The better you know a problem, the longer you wait before reviewing it.

Daily Selection Algorithm

The system creates a balanced daily practice set using these ratios:

Selection Ratios

For a typical 5-problem daily session:

  • 50% New (Gray): 2-3 problems

    • Keeps you learning new patterns
    • Expands your problem repertoire
  • 40% Review (Orange/Yellow): 2 problems

    • Reinforces recent learning
    • Prevents forgetting
  • 10% Mastered (Green): 0-1 problem

    • Maintains long-term retention
    • Keeps skills sharp

Example Daily Selection (5 problems)

code
Day 1:
- 2 Gray (new)
- 2 Orange (review from 3+ days ago)
- 1 Yellow (review from 7+ days ago)

Day 15:
- 3 Gray (new)
- 1 Orange (review)
- 1 Green (maintenance from 14+ days ago)

Optimization Strategies

When You Have Limited Time

3 problems/day (minimum effective dose):

  • 2 Gray (new)
  • 1 Orange/Yellow (review)

5 problems/day (recommended):

  • 3 Gray (new)
  • 2 Orange/Yellow (review)

7+ problems/day (intensive prep):

  • 3-4 Gray (new)
  • 2-3 Orange/Yellow (review)
  • 1 Green (maintenance)

When You're Behind on Reviews

If you have many eligible reviews piling up:

  1. Prioritize older reviews: Problems waiting longest get selected first
  2. Increase daily problem count: Temporarily do 7-10 problems/day
  3. Be honest with colors: If you can't solve it, mark it appropriately

When You're Running Out of New Problems

  1. Import more problems: Add problems from specific companies or topics
  2. Focus on reviews: Increase review ratio temporarily (30% new, 70% review)
  3. Reset old greens: Move ancient green problems back to yellow for refresh

How Colors Should Be Assigned

Be honest and strict with yourself:

Gray → Orange

  • You solved it with the right approach
  • May have needed hints on implementation details
  • Understood the solution when complete

Orange → Yellow

  • Solved independently without hints
  • Identified the correct pattern
  • May have been slower than optimal

Yellow → Green

  • Solved efficiently in reasonable time
  • Optimal time/space complexity
  • Can explain the solution clearly
  • Would be confident solving in an interview

Regression (Move Back)

  • Couldn't remember the approach
  • Needed to look up the solution
  • Got the wrong pattern entirely
  • Took much longer than expected

Advanced Techniques

Pattern-Based Learning

Group problems by pattern and learn them together:

  • All sliding window problems in a week
  • Then all DP problems the next week
  • This reinforces pattern recognition

Interleaving

Mix different patterns in daily practice:

  • Better for long-term retention
  • Improves pattern recognition
  • Mirrors real interview conditions

Deliberate Difficulty

If a color feels too easy:

  • Challenge yourself with harder variants
  • Set time limits (20-30 minutes)
  • Explain solution without code first

Common Mistakes to Avoid

1. Moving Colors Too Quickly

  • Don't mark something green just because you solved it once
  • Wait the full intervals between attempts
  • Be honest about your confidence level

2. Ignoring Reviews

  • Don't skip review problems to do only new ones
  • Reviews are critical for retention
  • Balance is key

3. Gaming the System

  • Don't look up solutions and mark as complete
  • This defeats the purpose of spaced repetition
  • You'll struggle in actual interviews

4. Not Adapting the Schedule

  • Adjust daily problem count based on available time
  • Scale ratios if you're short on new problems
  • Increase reviews before interviews

Measuring Progress

Healthy Metrics

  • Green percentage increasing: More problems mastered over time
  • Review success rate high: 80%+ of reviews completed without regression
  • Pattern recognition improving: Identifying patterns faster
  • Time per problem decreasing: Getting more efficient

Warning Signs

  • Too many eligible reviews: You're adding problems too fast
  • Frequent regressions: Need to wait longer between attempts
  • Everything is gray/orange: Not enough consistent practice
  • Everything is green: May need harder problems or be marking too generously

Integration with Interview Prep Timeline

3 Months Before Interview

  • Focus on new problems (60% new, 40% review)
  • Build broad pattern knowledge
  • Daily quota: 5-7 problems

1 Month Before Interview

  • Balanced approach (50% new, 50% review)
  • Solidify weak patterns
  • Daily quota: 5-7 problems

2 Weeks Before Interview

  • Review-heavy (30% new, 70% review)
  • Focus on company-specific patterns
  • Daily quota: 5-10 problems

1 Week Before Interview

  • Maintenance mode (20% new, 80% review)
  • Focus on greens and yellows
  • Keep skills sharp
  • Daily quota: 3-5 problems

Tips for Maximum Effectiveness

  1. Consistency > Intensity: Better to do 3 problems daily than 20 problems once a week
  2. Track patterns: Notice which patterns you struggle with
  3. Time yourself: Simulate interview pressure
  4. Write clean code: Practice interview-quality code, not just working code
  5. Explain out loud: Practice verbalizing your approach
  6. Review mistakes: Spend time understanding why you got stuck

The Science Behind It

Spaced repetition leverages:

  • Spacing Effect: Information is better retained when study sessions are spaced out
  • Testing Effect: Active recall strengthens memory more than passive review
  • Desirable Difficulty: Slightly challenging reviews optimize learning

The goal is to review each problem just before you would forget it - maximizing retention while minimizing review time.