Semrush MCP Server

An MCP (Model Context Protocol) server that provides access to Semrush API functionality for AI assistants.

Features

• Domain Overview: Get comprehensive domain analytics including organic traffic, keywords, and authority score

• Keyword Research: Analyze keyword metrics including search volume, difficulty, and CPC

• Organic Search Analysis: Retrieve organic search keywords and rankings for any domain

• Paid Search Analysis: Get Google Ads keywords and PPC data for domains

• Backlinks Overview: Get backlink metrics and analysis for domains or specific URLs

• Competitor Research: Identify and analyze organic competitors

• Related Keywords: Discover related keywords and search suggestions

Related MCP server: SEO AI Assistant

Prerequisites

• Node.js 18+

• A Semrush API key (get one at https://www.semrush.com/api/)

Installation

1. Clone this repository or create a new directory:

2. Create the following directory structure:

3. Save the provided files in their respective locations

Then manually:

Configuration

Set your Semrush API key as an environment variable:

Or create a .env file:

Usage with Claude Desktop

Add the server to your Claude Desktop configuration file (claude_desktop_config.json):

Available Tools

1. domain_overview

Get comprehensive domain analytics.

Parameters:

• domain (required): Domain to analyze (e.g., "example.com")

• database (optional): Database code (default: "us")

2. keyword_overview

Get keyword metrics and data.

Parameters:

• phrase (required): Keyword phrase to analyze

• database (optional): Database code (default: "us")

3. domain_organic_search

Get organic search keywords for a domain.

Parameters:

• domain (required): Domain to analyze

• database (optional): Database code (default: "us")

• limit (optional): Number of results (default: 10)

• offset (optional): Offset for pagination (default: 0)

4. backlinks_overview

Get backlinks overview for a domain or URL.

Parameters:

• target (required): Domain or URL to analyze

• target_type (optional): "domain" or "url" (default: "domain")

5. competitor_research

Find organic competitors for a domain.

Parameters:

• domain (required): Domain to analyze

• database (optional): Database code (default: "us")

• limit (optional): Number of competitors (default: 10)

6. domain_adwords

Get paid search (Google Ads) keywords for a domain.

Parameters:

• domain (required): Domain to analyze

• database (optional): Database code (default: "us")

• limit (optional): Number of results (default: 10)

7. related_keywords

Get related keywords and suggestions for a seed keyword.

Parameters:

• phrase (required): Seed keyword phrase

• database (optional): Database code (default: "us")

• limit (optional): Number of results (default: 10)

Database Codes

Common database codes for different regions:

• us - United States

• uk - United Kingdom

• ca - Canada

• au - Australia

• de - Germany

• fr - France

• es - Spain

• it - Italy

• br - Brazil

• in - India

Development

To run the server in development mode:

Example Usage in Claude

Once configured, you can use natural language to access Semrush data:

• "Analyze the domain example.com using Semrush"

• "What's the search volume for 'digital marketing' keyword?"

• "Show me the top organic keywords for techcrunch.com"

• "Find competitors for shopify.com"

• "Get backlink data for https://metehan.ai/blog/semrush-mcp/"

• "What related keywords can you find for 'content marketing'?"

• "Show me the Google Ads keywords that amazon.com is bidding on"

Troubleshooting

1. API Key Issues: Ensure your SEMRUSH_API_KEY environment variable is set correctly

2. Rate Limits: Semrush API has rate limits. Check your plan's limits

3. Database Codes: Make sure you're using valid database codes for your target regions

License

MIT