MCP Podcast Scraper MCP Server

MCP Podcast Scraper

An MCP (Model Context Protocol) server that scrapes and transcribes podcast episodes. Designed to work with Claude Code or Claude Desktop - you provide the podcast, the MCP transcribes it, and Claude summarizes it.

What It Does

• 🎙️ Scrapes podcasts from YouTube videos or RSS feeds
• 🎯 Transcribes audio using Deepgram's fast Nova-2 model
• 📁 Organizes files by podcast name and episode date
• 🔄 Tracks podcasts for new episodes
• ⏭️ Skips duplicates - won't re-scrape already processed episodes
• 📋 Finds incomplete work - lists episodes that need summarization
• ✏️ Custom summary prompts - customize how Claude summarizes for your needs

How It Works

You: "Check for new episodes and summarize them"
         ↓
Claude: Calls check_new_episodes() → Finds new episodes
         ↓
Claude: Calls scrape_podcast() → Downloads & transcribes
         ↓
Claude: Calls get_summary_prompt() → Reads your custom instructions
         ↓
Claude: Calls get_transcript() → Reads the transcript
         ↓
Claude: Summarizes following your prompt
         ↓
Claude: Calls save_summary() → Saves the .md file
         ↓
Done! transcript.md + summary.md saved

Installation Guide

Step 1: Prerequisites

Install required system tools (macOS):

# Install Homebrew if you don't have it
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

# Install yt-dlp (for YouTube) and ffmpeg (for audio)
brew install yt-dlp ffmpeg

Step 2: Clone & Build

# Clone the repository
git clone https://github.com/wkoleilat-happytitan/mcp-podcast-scraper.git
cd mcp-podcast-scraper

# Install dependencies
npm install

# Build
npm run build

Step 3: Get a Deepgram API Key

1. Go to https://console.deepgram.com/
2. Sign up (free tier includes $200 credit - enough for ~300 hours of audio)
3. Create an API key
4. Copy the key

Step 4: Configure

Copy the example config file and add your API key:

# Copy the example config
cp config.example.json config.json

# Edit config.json and add your Deepgram API key

Your config.json should look like:

{
  "outputDirectory": "./podcasts",
  "deepgramApiKey": "YOUR_ACTUAL_DEEPGRAM_API_KEY",
  "tempDirectory": "./temp"
}

⚠️ Important: Never commit config.json to git - it contains your API key! The .gitignore already excludes it.

Step 5: Add to Claude Code

Add this to your Claude Code MCP settings (~/.cursor/mcp.json or via Settings → MCP):

{
  "mcpServers": {
    "podcast-scraper": {
      "command": "node",
      "args": ["/FULL/PATH/TO/mcp-podcast-scraper/dist/index.js"]
    }
  }
}

Important: Replace /FULL/PATH/TO/ with the actual path to your installation.

Step 5 (Alternative): Add to Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "podcast-scraper": {
      "command": "node",
      "args": ["/FULL/PATH/TO/mcp-podcast-scraper/dist/index.js"]
    }
  }
}

Then restart Claude Desktop.

File Structure

mcp-podcast-scraper/
├── config.example.json     # Template - copy to config.json
├── config.json             # Your config (git-ignored, contains API key)
├── tracking.example.json   # Example tracking file
├── tracking.json           # Your tracked podcasts (git-ignored)
├── prompts/
│   └── summary-prompt.md   # Customize how Claude summarizes (editable)
├── podcasts/               # Your transcripts & summaries (git-ignored)
├── src/                    # Source code
├── dist/                   # Compiled code (git-ignored)
└── node_modules/           # Dependencies (git-ignored)

Usage Examples

Scrape a Specific Episode

"Scrape this YouTube podcast: https://youtube.com/watch?v=..."

"Find and scrape the latest Lex Fridman episode"

Track Podcasts for New Episodes

"Track the Huberman Lab podcast: https://feeds.megaphone.fm/hubermanlab"

"Check my tracked podcasts for new episodes"

"List all podcasts I'm tracking"

Find Incomplete Work

"Show me episodes that need summaries"

"List incomplete episodes"

MCP Tools Reference