yves-lou/mcp-brand-analyzer

Brand Analyzer MCP Server

A custom MCP (Model Context Protocol) server that performs deep brand analysis for personalized outreach with intelligent context-based email selection.

Features

• 🔍 Instagram Profile Scraping: Uses Apify to get real follower counts, bios, and website links
• 🌐 Website Analysis: Extracts brand voice, product information, and e-commerce indicators
• 📧 Intelligent Email Selection: Context-aware email extraction that analyzes surrounding text
• 📸 Image Quality Assessment: Analyzes product photography quality (resolution, lighting, composition)
• 🎯 Smart Service Recommendations: Intelligently matches brands to appropriate services
• ✉️ Natural Email Templates: Generates conversational, non-pushy outreach emails

Installation

cd mcp-brand-analyzer
npm install

Configuration

Environment Variables

Copy .env.example to .env and fill in your credentials:

cp .env.example .env

Required variables:

• APIFY_API_KEY: Your Apify API key from https://console.apify.com/account/integrations

Optional:

• PORT: Server port (default: 3100)

Configuration

Option 1: Use with Claude Desktop

Add to your Claude Desktop config (%APPDATA%\Claude\claude_desktop_config.json):

{
  "mcpServers": {
    "brand-analyzer": {
      "command": "node",
      "args": ["E:\\LEADS\\mcp-brand-analyzer\\index.js"]
    }
  }
}

Option 2: Use with n8n

See n8n-workflow-with-mcp.json for the updated workflow that uses this MCP server.

Tools

analyze_brand

Comprehensive brand analysis combining website + Instagram data

Input:

{
  "website": "https://example.com",
  "instagram": {
    "username": "brandname",
    "followers": 5000,
    "bio": "...",
    "latestPostCaption": "...",
    "engagement": 2.5,
    "videoPercentage": 20,
    "emojiPercentage": 15
  },
  "contactName": "Jane Doe",
  "location": "Paris, France"
}

Output:

• Website analysis
• Image quality scores
• Service recommendation with reasoning
• Pre-generated personalized email

analyze_website

Standalone website analysis

analyze_image_quality

Standalone image quality assessment

Service Recommendation Logic

1. Subtle AI Motion: High-quality photos + low video usage + e-commerce
2. Creative AI Imagery: Playful brand voice + emoji usage
3. Full Production/Video: High video usage + established following
4. Photography Consultation: Low image quality + e-commerce
5. E-commerce Animations: Moderate quality + online store

Why This is Better

Before (direct Claude API):

• Generic prompts with limited context
• No actual website/image analysis
• "Salesy" tone
• Low accuracy

After (MCP Server):

• Analyzes actual visual content
• Examines real product images
• Natural, conversational emails
• Smart service matching based on real data

Testing

node index.js

The server runs on stdio and communicates via JSON-RPC.