LearnFlow AI
AI-powered MCP server that transforms learning by finding best YouTube tutorials, generating personalized learning paths, and tracking progress for any tech skill.
README
<p align="center"> <img src="https://img.shields.io/badge/🚀-LearnFlow_AI-blueviolet?style=for-the-badge&labelColor=black" alt="LearnFlow AI"/> </p>
<h1 align="center">🎓 LearnFlow AI</h1> <h3 align="center">Your AI-Powered Learning Companion | Learn. Build. Play. 🚀</h3>
<p align="center"> <img src="https://img.shields.io/badge/License-MIT-green.svg" alt="License"/> <img src="https://img.shields.io/badge/Node.js-18+-brightgreen.svg" alt="Node"/> <img src="https://img.shields.io/badge/MCP-Compatible-blue.svg" alt="MCP"/> <img src="https://img.shields.io/badge/YouTube_API-v3-red.svg" alt="YouTube"/> <img src="https://img.shields.io/badge/AI_Powered-Claude-orange.svg" alt="Claude"/> </p>
<p align="center"> <b>Free • Open Source • Learn Anything • No Limits</b> </p>
🌟 What is LearnFlow AI?
LearnFlow AI is a revolutionary MCP (Model Context Protocol) server that transforms how you learn programming and technology. It connects directly to Claude AI and gives it superpowers to:
- 🎯 Find the BEST YouTube tutorials for any topic
- 📚 Generate personalized learning paths for ANY skill
- 🔍 Filter videos by quality (80%+ positive comments only!)
- ⏱️ Create time-based study plans (1 hour, weekend, daily micro-learning)
- 🎓 Track your learning progress
- 🆕 Stay updated with latest tech news and features
"Learn anything, build everything, pay nothing." 💪
🎬 Why LearnFlow AI?
| Problem | LearnFlow AI Solution |
|---|---|
| 😵 Overwhelmed by YouTube tutorials | 🎯 AI finds the BEST videos, filters out junk |
| 📚 Don't know where to start | 🗺️ Auto-generated learning roadmaps |
| ⏰ No time to learn | ⚡ 15-min daily micro-learning plans |
| 💸 Can't afford courses | 🆓 100% free YouTube-based learning |
| 📅 No structure | 📋 Week-by-week curriculum with projects |
| 🤔 Outdated content | 🆕 Finds recent, up-to-date tutorials |
🚀 Quick Start (5 Minutes Setup!)
Prerequisites
- Node.js 18+ - Download here
- YouTube Data API Key - Get free key here
- Claude Desktop - Download here
Step 1: Clone & Install
# Clone the repository
git clone https://github.com/Shalin-Shah-2002/Yt-MCP.git
# Navigate to project
cd Yt-MCP
# Install dependencies
npm install
Step 2: Get YouTube API Key (Free!)
- Go to Google Cloud Console
- Create a new project (or select existing)
- Enable YouTube Data API v3
- Go to Credentials → Create Credentials → API Key
- Copy your API key
Step 3: Configure Claude Desktop
Add this to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"learnflow-ai": {
"command": "node",
"args": ["/FULL/PATH/TO/Yt-MCP/src/server.js"],
"env": {
"YOUTUBE_API_KEY": "YOUR_API_KEY_HERE"
}
}
}
}
💡 Replace
/FULL/PATH/TO/Yt-MCPwith actual path and add your API key!
Step 4: Restart Claude Desktop
Close and reopen Claude Desktop. You should see LearnFlow AI tools available! 🎉
🛠️ All Features & Tools (27 Total!)
📹 YouTube Tools (14 Tools)
| Tool | Description |
|---|---|
search_youtube |
Search YouTube videos with smart ranking |
get_video_details |
Get comprehensive video info |
analyze_video |
Deep analysis with quality score & recommendations |
get_video_comments |
Analyze comment sentiment |
get_video_timestamps |
Extract chapters/timestamps |
compare_videos |
Side-by-side video comparison |
get_related_videos |
Find similar content |
get_video_transcript |
Get video transcript/summary info |
get_full_transcript |
Fetch complete video transcript |
analyze_channel |
Deep channel analysis |
get_channel_info |
Basic channel information |
set_youtube_api_key |
Set API key at runtime |
get_api_key_status |
Check API key status |
check_api_quota |
Monitor API usage |
🎓 Learning Path Tools (3 Tools)
| Tool | Description |
|---|---|
generate_learning_path |
Create week-by-week curriculum for ANY topic |
get_available_paths |
See suggested learning categories |
get_path_info |
Get details about a learning path |
🧠 Smart Learning Tools (10 Tools)
| Tool | Description |
|---|---|
smart_search |
Quality-filtered search (80%+ positive comments) |
generate_smart_roadmap |
AI-powered learning roadmap |
get_video_trust_score |
Calculate video quality score |
generate_topic_curriculum |
Detailed curriculum with phases |
get_prerequisites |
What to learn before a topic |
get_next_steps |
What to learn after completing topic |
find_best_video |
Find single best video for a topic |
get_learning_progress |
Track your learning journey |
update_learning_progress |
Mark topics as complete |
get_trending_in_tech |
Discover trending tech topics |
💬 18 Built-in Prompts for Learning
🎯 Quick Learning Prompts
| Prompt | What It Does |
|---|---|
find-best-recent |
Find BEST and most RECENT videos on any topic |
learn-in-one-hour |
Learn any topic in 60 minutes |
learn-before-tomorrow |
Emergency crash course for deadlines |
weekend-deep-dive |
Master a topic over the weekend (6-10 hrs) |
daily-micro-learning |
15-20 min daily learning habit |
📚 Structured Learning Prompts
| Prompt | What It Does |
|---|---|
create-learning-journey |
Personalized multi-week learning path |
weekly-study-plan |
Detailed plan for one week |
skill-assessment |
Assess your current level |
continue-learning |
Pick up where you left off |
quick-refresh |
Quick refresher on known topics |
🔍 Research & Discovery Prompts
| Prompt | What It Does |
|---|---|
deep-dive-topic |
Comprehensive research on a topic |
compare-technologies |
Compare frameworks/languages |
whats-new-in |
Latest updates in any technology |
explain-like-im-five |
Simple explanations for complex topics |
🛠️ Project & Career Prompts
| Prompt | What It Does |
|---|---|
project-roadmap |
Learn by building real projects |
code-along-project |
Find best build-along tutorials |
interview-prep |
Prepare for technical interviews |
debug-learning |
Get help when stuck |
📖 How to Use LearnFlow AI
Example 1: Learn React in One Hour
Just ask Claude:
Use the learn-in-one-hour prompt with topic="React hooks"
Example 2: Weekend Python Mastery
Use the weekend-deep-dive prompt with topic="Python" total_hours="8" include_project="yes"
Example 3: Find Best Recent Tutorials
Use find-best-recent prompt with topic="Next.js 14" recency="this-month"
Example 4: Generate Full Learning Path
Use generate_learning_path tool with goal="Machine Learning" level="beginner"
Example 5: Emergency Interview Prep
Use learn-before-tomorrow prompt with topic="System Design" purpose="interview" hours_available="4"
Example 6: Daily Learning Habit
Use daily-micro-learning prompt with topic="TypeScript" daily_minutes="20" days="7"
🎯 Learning Paths for Any Topic!
LearnFlow AI can generate learning paths for ANY topic. The system is fully dynamic!
Suggested Categories:
- 🌐 Web Development
- 📱 Mobile Development
- 🤖 Machine Learning
- 📊 Data Science
- ⚙️ DevOps
- ☁️ Cloud Computing
- 🔒 Cybersecurity
- 🎮 Game Development
- ⛓️ Blockchain
- 🎨 UI/UX Design
Popular Technologies:
React, Vue, Angular, Next.js, Node.js, TypeScript, Flutter, React Native, Swift, Kotlin, Python, TensorFlow, PyTorch, Docker, Kubernetes, AWS, Azure, and many more!
Three Skill Levels:
- 🌱 Beginner (6 weeks) - Start from zero
- 🌿 Intermediate (6 weeks) - Level up your skills
- 🌳 Advanced (8 weeks) - Become an expert
🏗️ Project Architecture
Yt-MCP/
├── src/
│ ├── server.js # MCP server entry point
│ ├── config/
│ │ └── config.js # Configuration management
│ ├── controllers/
│ │ ├── youtubeController.js
│ │ ├── learningPathController.js
│ │ └── smartLearningController.js
│ ├── models/
│ │ ├── TopicTree.js # Dynamic topic suggestions
│ │ └── LearningPath.js # Learning path structure
│ ├── services/
│ │ ├── youtubeService.js # YouTube API integration
│ │ ├── learningPathService.js
│ │ └── smartLearningService.js
│ ├── prompts/
│ │ └── promptManager.js # 18 built-in prompts
│ └── utils/
│ ├── videoFormatter.js
│ └── errorHandler.js
├── package.json
└── README.md
🔧 Configuration Options
Environment Variables
| Variable | Description | Required |
|---|---|---|
YOUTUBE_API_KEY |
Your YouTube Data API key | Yes |
Runtime Configuration
You can also set the API key through Claude:
Use set_youtube_api_key tool with apiKey="YOUR_KEY"
Check your API status:
Use get_api_key_status tool
🎮 Fun Ways to Use LearnFlow AI
🏆 Challenge Mode
"I want to learn Docker in one weekend. Create an intensive plan with a project!"
📅 Daily Streak
"Set up a 30-day JavaScript challenge with 20 minutes daily learning"
🆚 Tech Battle
"Compare React vs Vue vs Angular for a beginner who wants to get a job quickly"
🚀 Speed Run
"I have an interview tomorrow about Kubernetes. Emergency crash course please!"
🎯 Skill Tree
"Create a complete Full Stack Developer roadmap from zero to hero"
🔧 Development
Run in Development Mode
npm run dev
Run Tests
npm test
Check Syntax
node --check src/server.js
Test Individual Components
# Test transcript service
node test-transcript.mjs
# Test prompts
node -e "import('./src/prompts/promptManager.js').then(m => console.log(new m.PromptManager().listPrompts()))"
🌟 What Makes LearnFlow AI Special?
✨ Quality First
Only recommends videos with 80%+ positive comments. No more wasting time on bad tutorials!
🎯 Personalized
Adapts to YOUR skill level, time availability, and learning style.
🆓 Completely Free
Uses free YouTube content + free YouTube API. No subscriptions, no hidden costs.
🤖 AI-Powered
Leverages Claude AI to understand your needs and create perfect learning plans.
📱 Any Topic
Not limited to pre-defined paths. Learn literally anything from Kubernetes to Origami!
⏱️ Time-Aware
Whether you have 15 minutes or a whole weekend, LearnFlow creates the perfect plan.
📊 API Quota Information
YouTube Data API has a free quota of 10,000 units/day. Typical usage:
| Action | Cost |
|---|---|
| Search | 100 units |
| Video details | 1 unit |
| Comments | 1 unit |
This means you can do approximately 100 searches per day for free!
🐛 Troubleshooting
"API Key not configured"
Make sure you've set YOUTUBE_API_KEY in your Claude Desktop config or use the set_youtube_api_key tool.
"Claude doesn't show LearnFlow tools"
- Check the config file path is correct
- Restart Claude Desktop completely
- Check the path to
server.jsis absolute
"Quota exceeded"
You've hit the daily YouTube API limit. Wait 24 hours or create a new API key.
🗺️ Roadmap
- [ ] 📊 Learning analytics dashboard
- [ ] 🔄 Playlist support
- [ ] 📝 Note-taking integration
- [ ] 🎯 Quiz generation from videos
- [ ] 📱 Mobile companion app
- [ ] 🌍 Multi-language support
🤝 Contributing
We welcome contributions! Feel free to:
- 🐛 Report bugs
- 💡 Suggest features
- 🔧 Submit PRs
- ⭐ Star the repo
How to Contribute
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
📄 License
MIT License - Use it, modify it, share it! 🎉
🙏 Acknowledgments
- YouTube Data API - For access to the world's largest video library
- Anthropic Claude - For the amazing MCP protocol
- Open Source Community - For inspiration and support
<p align="center"> <b>🎓 Start Learning Today!</b><br><br> <i>"The best time to learn was yesterday. The second best time is now."</i> </p>
<p align="center"> <br> Made with ❤️ by <a href="https://github.com/Shalin-Shah-2002">Shalin Shah</a> </p>
<p align="center"> <br> <a href="https://github.com/Shalin-Shah-2002/Yt-MCP">⭐ Star this repo if it helps you learn something new!</a> </p>
<p align="center"> <b>🚀 Let's Learn Together! 🚀</b> </p>
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。