Semrush MCP

The Semrush MCP server provides AI agents with secure access to Semrush's public APIs.

Specifically, data from the following Semrush APIs is available via Semrush MCP:

• All Trends API (depending on your Trends API subscription level)
• All Analytics API v3
• All read-only methods from Projects API v3. That is, you can get data from your projects this way, but not create new projects, modify settings, or perform other actions.

The Semrush MCP server lets you integrate Semrush data into AI tools such as:

• Claude (browser and desktop)
• Claude Code
• Cursor
• VS Code

If you are interested in connecting Semrush data to other AI agents and LLM tools through MCP, reach out to Semrush Support. Each request will be reviewed individually.

With Semrush MCP, you can pull Semrush data through your AI agents and speed up decision-making. For example:

• Scan keyword and backlink data daily with an AI agent to spot ranking drops and growth opportunities early.
• Track competitor traffic and get alerts when their performance changes.
• Generate monthly traffic and SEO reports automatically in Google Docs or Notion to save time on manual reporting.
• Add Semrush data directly to dashboards, reports, or analytics features.

No more custom builds. Just one connection that brings Semrush insights straight into your AI workflows.

Once you connect an AI tool like Claude to Semrush MCP, you can write a prompt such as “List domains in the human resources industry for July 2025.”

Claude would connect to the Semrush API via MCP, fetch and filter matching domains, and respond with a list that matches your query. This helps you find competitors in your industry.

You can also compare the performance of specific domains with a query like "Give me a traffic comparison for nike.com and adidas.com for July 2024 through July 2025." Your workflow would look like this:

Access to the Semrush MCP Server is included in the SEO Business plan (which provides Standard API access) and with any Trends API subscription (Basic or Premium). With either option, you can begin using AI integrations right away.

The Semrush MCP server uses the same unit-based system as the Semrush API to manage and limit API usage. Each API request consumes a certain number of units, depending on the type of report and the amount of data requested. To get an idea of how many API units a request needs, refer to:

• Semrush Trends API reference
• Semrush API limits consumption

Make sure your AI agent:

• Supports OAuth 2.1.
• Supports streamable HTTP Transport.

This server uses streamable HTTP transport only. The MCP endpoint is:

https://mcp.semrush.com/v1/mcp

Authentication is handled through OAuth 2.1, which will automatically redirect you to the Semrush login and consent pages when needed.

Different MCP clients require different configuration formats:

Cursor

In Settings → Cursor Settings → MCP & Integrations, add the Semrush MCP configuration to your Cursor MCP settings:

The OAuth 2.1 authentication process will automatically redirect you to the Semrush login page when needed.

For more details, refer to the official Cursor MCP documentation.

Claude

To add the Semrush MCP server to Claude (browser or desktop):

1. Navigate to Settings → Connectors → Add custom connector.
2. Name the connector and add the Semrush MCP server URL: https://mcp.semrush.com/v1/mcp
3. Save the changes.

For more information, refer to the official Anthropic documentation.

Claude Code

To add the Semrush MCP endpoint to your Claude Code settings, add a remote HTTP server by running the following command:

For more details, refer to the official Claude MCP documentation.

VS Code

To add the Semrush MCP endpoint to your VS Code settings:

1. Create a .vscode/mcp.json file in your VS Code workspace.
2. Add the following MCP configuration to the file and save the changes:

The OAuth 2.1 authentication process will automatically redirect you to the Semrush login page when needed.

For more details, refer to the official VS Code MCP documentation.

Troubleshooting

OAuth authentication problems

If you're getting the redirect URI <AI_AGENT_URI> is not allowed error, contact Customer Support, sharing the AI agent name and redirect URI.

Connection issues

• Verify the Semrush MCP server URL is correct: https://mcp.semrush.com/v1/mcp.
• Check your client's MCP configuration syntax.

Issues querying data

• Confirm that you have access to the API you want to get data from based on your Semrush subscription. For example, you won't be able to get Trends API data unless you have the Trends API Basic or Premium. To understand what access you have, review your subscription in My Profile → Subscription info → Summary, SEO and/or Traffic & Market.
• Make sure you have enough API units to make calls. To see the Standard and Trends API units you have, check My Profile → Subscription info → Summary.

Missing tools

Your AI agent should be using the semrush_report, semrush_report_list, and semrush_report_schema to get and understand the Semrush data. If not all of them are available in your AI agent:

• Make sure that authentication is completed successfully.
• Verify your Semrush access.
• Check for any error messages in your AI agent’s console.