The DataForSEO MCP server is a powerful tool that allows the seamless integration of DataForSEO APIs and enables advanced SEO data analytics capabilities. This simple guide will walk you through the process of setting up and using the official DataForSEO MCP server.
What is DataForSEO MCP server?
DataForSEO MCP server is the implementation of the Model Context Protocol, allowing AI tools to obtain and process data from DataForSEO. In a nutshell, this MCP server is a bridge between AI models and our APIs, so that you can easily extract data and turn it into insights without writing a single line of code.
This article will cover the following:
Quick start: connect from your AI tool or client
Run without installation
Advanced setup: run from source
Streamable HTTP
FAQ & troubleshooting
Quick start: connect from your AI tool or client
Configuring and using DataForSEO MCP server is really easy. You can connect it to various AI models and tools by following our quick setup guides for:
➤ Claude (Desktop)
➤ Claude Code
➤ Gemini CLI
➤ ChatGPT
➤ Docker
➤ Cursor
➤ DataForSEO MCP Server + AI + n8n (video)
Run without installation
The easiest way to run the DataForSEO MCP server is with npx. This way you always get the latest published version without needing a local installation.
Before getting started, make sure:
- You installed the latest version of Node.js.
- You have a DataForSEO account with your API login and password.
1 Set your environment variables. Remember to replace your_username and your_password with your actual DataForSEO API login and password.
macOS/Linux
export DATAFORSEO_USERNAME=your_username export DATAFORSEO_PASSWORD=your_password
Windows PowerShell
$env:DATAFORSEO_USERNAME = "your_username" $env:DATAFORSEO_PASSWORD = "your_password"
Windows Command Prompt
set DATAFORSEO_USERNAME="your_username" set DATAFORSEO_PASSWORD="your_password"
2 Run with npx to start the server using the stdio (standard input/output) protocol.
npx dataforseo-mcp-server
You can choose between different protocols depending on your client’s compatibility.
➤ HTTP server. Most modern MCP-compatible clients (except Claude Desktop) expect this protocol.
npx dataforseo-mcp-server http
➤ SSE (legacy protocol)
npx dataforseo-mcp-server sse
When you see DataForSEO MCP Server running on stdio, the server is ready.
Note that the DataForSEO MCP Server is just a backend. To actually use it, you need to connect it to an AI client like Claude Desktop, Cursor, or another MCP-compatible app.
Streamable HTTP
You can connect most supported AI clients (except Claude) to DataForSEO MCP server using a streamable HTTP protocol. There are several options:
Run it locally and then connect your client to the local URL.
npx dataforseo-mcp-server http
Use this URL (no local setup).
https://mcp.dataforseo.com/mcp
Advanced setup: run from source
If you prefer to run the server from source for development or offline use, follow the instructions in our GitHub repository. You can also check this video for details.
FAQ & troubleshooting
Auth errors (“you are not authorized…”)
This is the most common issue. Solutions:
- Make sure you’re using your API credentials (login and password) for making API calls, not your account login and password for logging into DataForSEO Dashboard.
- Note that the required format of API credentials depends on the client. Some tools only need you to enter your API login and API password into separate lines in the interface. ChatGPT with remote server requires an authorization header with
your_api_login:your_api_passwordencoded in the Base64 format like in this example: - Since each tool handles authentication differently, the safest approach is to follow the specific setup guide for your client.
Basic ZGVtb19sb2dpbjpkZW1vX3Bhc3N3b3Jk
AI model limits
Sometimes requests fail due to provider-side limits. Here are some workarounds we suggest:
- Ask the AI to wrap long responses into a file instead of outputting everything in the chat (this saves tokens).
- Start a new chat, reload the app, or wait until limits reset (for Claude, resets happen about every 5 hours).
- Temporarily disable unnecessary DataForSEO MCP tools to reduce token usage.
“Fetch not defined” error
This can be caused by old Node.js versions.
- Solution: install the latest Node.js version.