# Interleaved Learning MCP Server - Complete Documentation
> A Model Context Protocol (MCP) server implementing cognitive science-backed interleaved learning techniques for optimal knowledge retention.
## MCP Endpoint
https://interleaved-learning-mcp.vercel.app/api/mcp
## What is This?
This is a remote MCP (Model Context Protocol) server that provides AI assistants with tools for implementing interleaved learning - a research-backed cognitive strategy proven to improve long-term retention by up to 43% compared to traditional blocked practice.
## What is Interleaved Learning?
Interleaved learning involves mixing different topics during study sessions instead of focusing on one topic at a time. Research shows this approach:
- Improves long-term retention (up to 43% better - Rohrer & Taylor, 2007)
- Enhances ability to distinguish between similar concepts
- Improves transfer of knowledge to new situations
- Builds more flexible problem-solving skills
The key insight: The "desirable difficulty" of switching between topics forces the brain to practice retrieval and discrimination, strengthening neural pathways.
## Connection Details
### For Cursor IDE
Add to `.cursor/mcp.json`:
```json
{
"mcpServers": {
"interleaved-learning": {
"url": "https://interleaved-learning-mcp.vercel.app/api/mcp"
}
}
}
```
### For Claude Desktop
Add to `claude_desktop_config.json`:
```json
{
"mcpServers": {
"interleaved-learning": {
"transport": {
"type": "streamable-http",
"url": "https://interleaved-learning-mcp.vercel.app/api/mcp"
}
}
}
}
```
## Available Tools (7 Total)
### 1. create_study_plan
Generate an interleaved study schedule for multiple subjects.
**Parameters:**
- subjects (array, required): List of subjects to study (2-6 items)
- totalMinutes (number, required): Total study time (20-240 minutes)
- pattern (string, optional): Interleaving pattern to use
**Available Patterns:**
- ABAB: Simple alternation between two subjects (beginner)
- ABCABC: Triple rotation through three subjects (intermediate)
- ABACBC: Spaced mixing for better retention (intermediate)
- Random: Random shuffle for maximum interleaving (advanced)
- Blocked-to-Interleaved: Start blocked, gradually increase mixing (beginner)
**Example:**
```json
{
"subjects": ["Mathematics", "Physics", "Chemistry"],
"totalMinutes": 90,
"pattern": "ABCABC"
}
```
**Returns:** Structured study plan with time blocks, activities, and tips.
---
### 2. generate_interleaved_quiz
Create a mixed-topic quiz that strengthens discrimination learning.
**Parameters:**
- topics (array, required): Array of topics with questions
- name (string): Topic name
- questions (array): Questions with question, options, correctIndex
- questionsPerTopic (number, optional): Questions per topic (default: 3)
**Example:**
```json
{
"topics": [
{
"name": "Algebra",
"questions": [
{
"question": "Solve: 2x + 5 = 15",
"options": ["x = 5", "x = 10", "x = 7.5", "x = 3"],
"correctIndex": 0
}
]
}
],
"questionsPerTopic": 3
}
```
**Returns:** Shuffled quiz with questions mixed across topics.
---
### 3. create_flashcard_deck
Create a multi-topic flashcard deck for interleaved review.
**Parameters:**
- deckName (string, required): Name for the deck
- cards (array, required): Flashcards with front, back, and topic
**Example:**
```json
{
"deckName": "Science Review",
"cards": [
{ "front": "What is photosynthesis?", "back": "Process by which plants convert light to energy", "topic": "Biology" },
{ "front": "F = ma represents?", "back": "Newton's Second Law of Motion", "topic": "Physics" }
]
}
```
**Returns:** Confirmation with deck statistics and topic breakdown.
---
### 4. get_shuffled_flashcards
Retrieve flashcards in shuffled (interleaved) order for study.
**Parameters:**
- deckName (string, required): Name of the deck to retrieve
- count (number, optional): Number of cards to get (default: all)
**Example:**
```json
{
"deckName": "Science Review",
"count": 10
}
```
**Returns:** Shuffled flashcards across all topics with study tips.
---
### 5. log_study_session
Record a completed study session for progress tracking.
**Parameters:**
- userId (string, required): User identifier
- subject (string, required): Subject studied
- duration (number, required): Duration in minutes
- quizScore (number, optional): Quiz score percentage (0-100)
**Example:**
```json
{
"userId": "student123",
"subject": "Mathematics",
"duration": 45,
"quizScore": 85
}
```
**Returns:** Confirmation with session details and cumulative stats.
---
### 6. get_learning_progress
View learning statistics and get personalized recommendations.
**Parameters:**
- userId (string, required): User identifier
**Example:**
```json
{
"userId": "student123"
}
```
**Returns:** Progress breakdown by subject with:
- Total study time and sessions
- Average quiz scores per subject
- Personalized recommendations for improvement
---
### 7. get_interleaving_patterns
Learn about available interleaving patterns and their benefits.
**Parameters:** None required
**Returns:** Detailed information about each pattern including:
- Pattern descriptions and examples
- Difficulty levels
- Best use cases
- Research-backed benefits
## Research Background
Interleaved learning is supported by extensive cognitive science research:
1. **Rohrer & Taylor (2007)**: Found 43% improvement in test performance with interleaving vs. blocking
2. **Kornell & Bjork (2008)**: Demonstrated benefits for learning artist painting styles
3. **Taylor & Rohrer (2010)**: Showed interleaving helps students learn to solve different types of math problems
Key mechanisms:
- **Discrimination learning**: Forces learners to identify which strategy applies
- **Retrieval practice**: Switching topics requires recalling previous material
- **Desirable difficulty**: Challenge during learning strengthens memory traces
## Technical Details
- **Protocol**: Model Context Protocol (MCP)
- **Transport**: Streamable HTTP
- **Framework**: Next.js 15
- **Adapter**: mcp-handler (Vercel)
- **Validation**: Zod schemas
- **Deployment**: Vercel Edge Functions
## Repository
https://github.com/sheikhcoders/interleaved-learning-mcp
## Author
Likhon Sheikh
GitHub: @sheikhcoders
Email: sheikhcoders@gmail.com
## License
MIT License © 2024 Likhon Sheikh
## Related Links
- MCP Specification: https://modelcontextprotocol.io/
- Vercel MCP Docs: https://vercel.com/docs/mcp
- llms.txt Standard: https://llmstxt.org/
