YouTube MCP Tools for VS Code

🎬 Bring YouTube video intelligence directly into VS Code
Search • Analyze • Transcribe • Generate Flashcards

✨ What it does

The transcript tools work without a Google API key. That covers the single most common use case ("summarize this video for me") with zero setup. Add an API key later if you want search and analysis too.

⚡ Quick start (MCP server, no API key)

The fastest way to try it from any MCP client (Claude Desktop, VS Code Agent Mode, Cursor, custom agents):

1. Add to your MCP client config

.vscode/mcp.json:

{
  "servers": {
    "youtube": {
      "command": "node",
      "args": ["${workspaceFolder}/node_modules/youtube-mcp-tools/dist/mcp-server.js"]
    }
  }
}

Claude Desktop (claude_desktop_config.json):

{
  "mcpServers": {
    "youtube": {
      "command": "node",
      "args": ["/absolute/path/to/dist/mcp-server.js"]
    }
  }
}

2. Ask your AI

"Get the transcript of https://youtu.be/jNQXAC9IVRw and tell me what it's about."

"In https://youtu.be/dQw4w9WgXcQ, where do they say 'never gonna'?"

The AI calls youtube_get_transcript or youtube_search_transcript, gets back Markdown with [03:42](https://youtu.be/X?t=222) clickable timestamps, and cites them in its answer.

3. (Optional) Add an API key for richer tools

Set YOUTUBE_API_KEY in your MCP client's env block to unlock search, video details, full analysis, and flashcards.

📦 Installation (VS Code extension)

1. Open VS Code
2. Press Ctrl+P / Cmd+P
3. Type ext install fabioc-aloha.youtube-mcp-tools
4. Press Enter

Or search for "YouTube MCP Tools" in the Extensions view.

⚙️ Setup

Get a YouTube API Key

1. Go to Google Cloud Console
2. Create a new project (or select existing)
3. Enable the YouTube Data API v3
4. Create an API Key
5. Copy the key

Configure the Extension

Option 1: VS Code Settings (Recommended)

Ctrl+, → Search "youtubeMcp.apiKey" → Paste your key

Option 2: .env File Create a .env file in your workspace root:

YOUTUBE_API_KEY=your-api-key-here

Option 3: Secure Command (Recommended)

Run YouTube MCP: Set API Key command. Your key is:

• Validated before storage
• Encrypted using VS Code SecretStorage
• Never exposed in settings or files

🎯 Commands

Access all commands via Ctrl+Shift+P / Cmd+Shift+P.

🔒 Security Features

• Encrypted Storage - API keys stored in VS Code's SecretStorage
• Key Validation - API keys validated before storage
• Rate Limiting - Automatic request throttling (10 req/sec)
• Exponential Backoff - Smart retries for transient failures
• Quota Persistence - Usage tracking persists across sessions

📊 Video Analysis Output

When you analyze a video, you get:

# Video Title

**Channel:** Channel Name
**Duration:** 15m 30s
**Views:** 1,234,567 | **Likes:** 45,678

---

## Summary
Brief overview of the video content...

### Detailed
In-depth analysis of the video...

## Key Points
- Important point 1
- Important point 2
- Important point 3

## Key Concepts
### Machine Learning
Definition and context...
*Type: term | Mentions: 15*

## Quality Signals
| Signal | Value |
|--------|-------|
| Has captions | yes |
| Word count | 4,318 |
| Avg sentence length | 18 words |
| Engagement (likes/views) | 3.70% |
| Transcript segments | 412 |

Quality is reported as measurable signals, not opaque 0–100 scores. Interpret them per your use case — long-form lectures want different signals than short tutorials.

🤖 Use From Any MCP Client

Beyond the VS Code UI, this package ships a standalone Model Context Protocol server (youtube-mcp-server). Any MCP-aware client can call its tools — youtube_search, youtube_get_video_details, youtube_get_transcript, youtube_analyze_video, youtube_generate_flashcards, youtube_quota_status.

VS Code Agent Mode (.vscode/mcp.json)

{
  "servers": {
    "youtube": {
      "command": "node",
      "args": ["${workspaceFolder}/node_modules/youtube-mcp-tools/dist/mcp-server.js"],
      "env": { "YOUTUBE_API_KEY": "${env:YOUTUBE_API_KEY}" }
    }
  }
}

Claude Desktop (claude_desktop_config.json)

{
  "mcpServers": {
    "youtube": {
      "command": "node",
      "args": ["/absolute/path/to/dist/mcp-server.js"],
      "env": { "YOUTUBE_API_KEY": "AIza..." }
    }
  }
}

Standalone

YOUTUBE_API_KEY=AIza... node dist/mcp-server.js

The server speaks JSON-RPC over stdio. Logs go to stderr only — stdout is reserved for the protocol channel.

🎴 Flashcard Generation

Flashcards are generated from:

• Key Points extracted from the video
• Concepts identified in the content
• Topics covered in the video

Perfect for:

• 📚 Learning new technologies
• 🎓 Studying for certifications
• 💡 Retaining video course content

⚙️ Settings

📁 Activity Bar Views

Look for the YouTube icon in the Activity Bar:

• 🔍 Search Results - Videos from your last search
• 📜 Recent Videos - Your video history
• 🎴 Flashcards - Generated study cards

🔒 Privacy & Security

• ✅ API keys stored locally in VS Code settings or .env
• ✅ No data sent to third-party servers
• ✅ No telemetry or tracking
• ✅ All processing happens locally

📝 Requirements

• VS Code 1.109.0 or higher
• YouTube Data API key (Get one free)

🤝 Contributing

Contributions welcome! Please see our GitHub repository.

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

Made with ❤️ by CorreaX
_{Building intelligent tools for developers}

Report Bug • Request Feature • ⭐ Star on GitHub