Table of Contents

• Features
• Installation
• Quick Start
• Tools
  • repurpose_content
  • generate_hooks
  • extract_script
  • multi_platform_pack
• Configuration
  • Claude Desktop
  • Cursor
  • Other MCP Clients
• Environment Variables
• API Reference
• Development
• Contributing
• License

Features

| Feature | Description | |---------|-------------| | 🚀 4 Powerful Tools | Repurpose content, generate hooks, extract scripts, multi-platform packs | | 📱 Multi-Platform | Instagram, LinkedIn, Twitter, YouTube support | | ⚡ High Performance | In-memory caching with 5-minute TTL | | 🔄 Retry Logic | Built-in retry mechanism for LLM API calls | | 📝 Type-Safe | Zod validation for all inputs | | 🏗️ ES Modules | Modern Node.js module system | | 📦 Zero Dependencies | Lightweight - only uses proven libraries | | 🎯 MCP Compliant | Full Model Context Protocol support |

Installation

Prerequisites

• Node.js 18 or higher
• npm or yarn

Install from npm

# Global installation
npm install -g content-mcp

# Or use npx without installing
npx content-mcp

Install from source

# Clone the repository
git clone https://github.com/RonakMunjapara/content-mcp.git
cd content-mcp

# Install dependencies
npm install

# Build the project
npm run build

Quick Start

CLI Usage

# Run in stdio mode (for Claude Desktop)
npm start

# Run in HTTP mode
npm run start:http

# Development mode with hot reload
npm run dev

HTTP Mode

# Start HTTP server on default port 8080
node dist/server.js --http

# Custom port
PORT=3000 node dist/server.js --http

# Test the server
curl http://localhost:8080/health

Tools

1. repurpose_content

Transforms content into platform-specific viral posts with optimized hooks, captions, and hashtags.

Input:

{
  "content": "Your original content here...",
  "platform": "instagram" | "linkedin" | "twitter" | "youtube"
}

Output:

{
  "caption": "Optimized caption for the platform",
  "hashtags": ["#relevant", "#hashtags"],
  "hook": "Engaging hook to grab attention"
}

Platform-Specific Formatting:

| Platform | Max Length | Tone | CTA Style | |----------|------------|------|-----------| | Instagram | 2,200 | Conversational, emoji-friendly | Comment prompt | | LinkedIn | 3,000 | Professional, thought-leadership | Discussion prompt | | Twitter | 280 | Concise, punchy | RT request | | YouTube | 5,000 | Exciting, curiosity-driven | Subscribe prompt |

2. generate_hooks

Generates 10 viral hooks using proven psychological patterns.

Input:

{
  "content": "Your content to generate hooks for..."
}

Output:

{
  "hooks": [
    "What nobody tells you about [topic]...",
    "The secret to [topic] nobody talks about...",
    "Stop struggling with [topic]. Here's the fix...",
    "This [topic] mistake costs you years...",
    "I spent years learning this about [topic]...",
    "What if [topic] is actually simpler?",
    "You're doing [topic] wrong. Here's why...",
    "The hidden truth about [topic]...",
    "Why most people fail at [topic]...",
    "Ready to transform your [topic] game?"
  ]
}

Hook Styles:

• 🔥 Curiosity - Pique interest with the unknown
• ⚡ Urgency - Create FOMO with time-sensitive language
• 😱 Shock - Use surprising statistics or revelations
• 😤 Pain - Address common frustrations
• 🎯 Result - Show measurable outcomes
• ❓ Question - Engage with direct questions

3. extract_script

Converts content into a 20-30 second reel/short script optimized for maximum engagement.

Input:

{
  "content": "Your long-form content..."
}

Output:

{
  "script": "[HOOK - 3 seconds]\nAttention-grabbing opening\n\n[BODY - 20 seconds]\nCore content with key points\n\n[CTA - 4 seconds]\nEngagement call-to-action"
}

Script Structure:

| Segment | Duration | Purpose | |---------|----------|---------| | Hook | 3 seconds | Grab attention immediately | | Body | 20 seconds | Deliver core value | | CTA | 4 seconds | Drive engagement |

4. multi_platform_pack

Generates optimized content for all platforms simultaneously.

Input:

{
  "content": "Your content to repurpose..."
}

Output:

{
  "instagram": {
    "caption": "...",
    "hashtags": ["..."]
  },
  "linkedin": {
    "post": "..."
  },
  "twitter": {
    "thread": ["tweet1", "tweet2", "..."]
  },
  "youtube": {
    "title": "...",
    "description": "..."
  }
}

Configuration

Claude Desktop

Method 1: Using npm start (Recommended)

{
  "mcpServers": {
    "content-repurposer": {
      "command": "npm",
      "args": ["start", "--prefix", "/path/to/content-mcp"],
      "env": {}
    }
  }
}

Method 2: Using node directly

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

Method 3: Using npx (After publishing)

{
  "mcpServers": {
    "content-repurposer": {
      "command": "npx",
      "args": ["content-mcp"],
      "env": {}
    }
  }
}

Config File Locations:

| OS | Path | |----|------| | macOS | ~/Library/Application Support/Claude/settings.json | | Windows | %APPDATA%\Claude\settings.json | | Linux | ~/.config/Claude/settings.json |

Cursor

{
  "mcpServers": {
    "content-repurposer": {
      "command": "npm",
      "args": ["start", "--prefix", "/path/to/content-mcp"],
      "env": {}
    }
  }
}

Other MCP Clients

Via HTTP

# Start the server in HTTP mode
node dist/server.js --http

# The server will be available at
http://localhost:8080

Via Stdio

# Pipe JSON-RPC commands to stdin
echo '{"jsonrpc":"2.0","method":"initialize","params":{...}}' | node dist/server.js

Environment Variables

| Variable | Default | Description | |----------|---------|-------------| | PORT | 8080 | HTTP server port | | NODE_ENV | development | Environment (development/production) |

API Reference

MCP Protocol

The server implements the full MCP protocol with these methods:

initialize

Initialize the server and get capabilities.

{
  "jsonrpc": "2.0",
  "id": "1",
  "method": "initialize",
  "params": {
    "protocolVersion": "2024-11-05",
    "capabilities": {},
    "clientInfo": {
      "name": "claude-desktop",
      "version": "1.0.0"
    }
  }
}

tools/list

List available tools.

{
  "jsonrpc": "2.0",
  "id": "2",
  "method": "tools/list",
  "params": {}
}

tools/call

Call a specific tool.

{
  "jsonrpc": "2.0",
  "id": "3",
  "method": "tools/call",
  "params": {
    "name": "repurpose_content",
    "arguments": {
      "content": "Your content...",
      "platform": "instagram"
    }
  }
}

Development

Project Structure

content-mcp/
├── src/
│   ├── server.ts          # Main MCP server
│   ├── utils/
│   │   ├── logger.ts       # Logging utility
│   │   ├── cache.ts        # In-memory cache
│   │   └── retry.ts       # Retry logic
│   └── tools/
│       ├── repurposeContent.ts
│       ├── generateHooks.ts
│       ├── extractScript.ts
│       └── multiPlatformPack.ts
├── package.json
├── tsconfig.json
├── README.md
└── RUN.md

Available Scripts

# Build TypeScript
npm run build

# Start server (stdio mode)
npm start

# Start server (HTTP mode)
npm run start:http

# Development with hot reload
npm run dev

# Type checking
npm run typecheck

Adding LLM Support

The retry utility is ready for LLM integration. To add OpenAI or Anthropic:

import { retry } from './utils/retry.js';
import axios from 'axios';

async function callLLM(prompt: string) {
  return retry(async () => {
    const response = await axios.post('https://api.openai.com/v1/chat/completions', {
      model: 'gpt-4',
      messages: [{ role: 'user', content: prompt }]
    }, {
      headers: { Authorization: `Bearer ${process.env.OPENAI_API_KEY}` }
    });
    return response.data;
  }, { maxAttempts: 3, delayMs: 1000 });
}

Contributing

Contributions are welcome! Please read our contributing guidelines first.

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

License

MIT License - see LICENSE for details.

Related

• Model Context Protocol Documentation
• MCP SDK
• Claude Desktop