DataForSEO SEO MCP Server · Model Context Protocol

Give your AI assistant real SEO data

The DataForSEO MCP server connects Claude, Cursor, and any MCP-compatible agent to seven live SEO APIs. Ask in plain English; get real SERPs, keywords, backlinks, and audits back.

Try for free

$1 in Free Credits · No Credit Card · No Subscription · Credits Never Expire

7 APIs exposed: SERP, Keywords, Labs, Backlinks, On-Page, Business Data, Domain Analytics
$0 extra cost for the MCP connection
1 command to run it: npx dataforseo-mcp-server
$1 free credit to start asking

The problem

Your AI is smart. Its SEO data is made up.

LLMs will happily invent search volumes and “analyze” rankings they’ve never seen. Grounding them in real data changes everything.

Hallucinated metrics

Ask a bare LLM for keyword volumes or a backlink profile and you’ll get confident numbers with no connection to reality.

→ MCP tools return live DataForSEO data your AI can cite.

Copy-paste research loops

Exporting from SEO tools, pasting into chats, re-formatting — the “AI workflow” that’s slower than doing it by hand.

→ The assistant queries the data itself, mid-conversation.

Automation still needs a developer

Every custom SEO workflow used to mean API integration work before you saw any value.

→ One config line connects Claude or Cursor — no code.

Connect your assistant free

1

Open Claude Desktop and go to Customize → Connectors.
2

Click Add custom connector and paste the DataForSEO MCP HTTP URL:
https://mcp.dataforseo.com/mcp
3

When prompted, approve the Authorization request to access the DataForSEO API.
4

Start a new chat. Claude will request your approval the first time it calls a DataForSEO tool.

1

Open the configuration file:
~/Library/Application Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the following configuration to the file:

JSON

{
  "mcpServers": {
    "dataforseo": {
      "command": "npx",
      "args": ["-y", "dataforseo-mcp-server"],
      "env": {
        "DATAFORSEO_USERNAME": "your_api_login",
        "DATAFORSEO_PASSWORD": "your_api_password"
      }
    }
  }
}

3

Replace your_api_login and your_api_password with credentials from the API Access dashboard.
4

Restart Claude Desktop. The DataForSEO tools will appear in your tool list.

1

Generate a Base64 token from your API login and password (or copy a pre-encoded one from the API Access dashboard via the Send by Email button).

Add the remote DataForSEO MCP server in your terminal:

Bash

claude mcp add --header "Authorization: Basic " --transport http dfs-mcp https://mcp.dataforseo.com/mcp

3
Verify the connection:
Bash
```
claude mcp list
```

1

Make sure you have Node.js 18+ installed.

Bash

claude mcp add dataforseo \
  --env DATAFORSEO_USERNAME=your_api_login \
  --env DATAFORSEO_PASSWORD=your_api_password \
  -- npx -y dataforseo-mcp-server

3
Verify the connection:
Bash
```
claude mcp list
```
Check our Help Center for more ways to install DataForSEO MCP with Claude Code.

1

In Cursor, open Settings → Tools & Integrations → MCP Tools.

Click Add Custom MCP and paste the remote configuration:

JSON

{
  "mcpServers": {
    "dataforseo": {
      "url": "https://mcp.dataforseo.com/mcp",
      "headers": {
        "Authorization": "Basic "
      }
    }
  }
}

3

Replace with your Base64-encoded username:password from the API Access dashboard.
4

Save and reopen the chat. Cursor supports up to 40 active MCP tools — disable any modules you don’t need.

1

Open the global Cursor MCP configuration:
~/.cursor/mcp.json
You can also access this file from Settings → Tools & Integrations → Add Custom MCP.

Add the local NPX configuration:

JSON

{
  "mcpServers": {
    "dataforseo": {
      "command": "npx",
      "args": ["-y", "dataforseo-mcp-server"],
      "env": {
        "DATAFORSEO_USERNAME": "your_api_login",
        "DATAFORSEO_PASSWORD": "your_api_password"
      }
    }
  }
}

3

Replace credentials with values from the API Access dashboard and save the file.
4

Restart Cursor or reload the MCP servers from Tools & Integrations.

1

Log into the OpenAI platform dashboard, go to Settings → Apps → Advanced Settings and enable the Developer mode.
2

Go back to Apps and click Create app.
3

Paste the DataForSEO MCP server URL and give the app a name:
https://mcp.dataforseo.com/mcp
4

Select the OAuth authentication method and click Create
5

When prompted, approve the Authorization request to access the DataForSEO API.
6

Start a new chat. ChatGPT will request your approval the first time it calls a DataForSEO tool.

1
Install the Gemini CLI globally:
Bash
```
npm install -g @google/gemini-cli
```
2

Open the Gemini settings file:
~/.gemini/settings.json
Windows: C:\Users\\.gemini\settings.json

Add the DataForSEO MCP server:

JSON

{
  "selectedAuthType": "gemini-api-key",
  "mcpServers": {
    "dataforseo": {
      "command": "npx",
      "args": ["-y", "dataforseo-mcp-server"],
      "env": {
        "DATAFORSEO_USERNAME": "your_api_login",
        "DATAFORSEO_PASSWORD": "your_api_password"
      }
    }
  }
}

4

Restart the CLI and run /mcp list to confirm DataForSEO is connected.

1

Install Docker for your operating system.

Pull and run the official DataForSEO MCP image:

Bash

docker run -p 3000:3000 \
  -e DATAFORSEO_USERNAME=your_api_login \
  -e DATAFORSEO_PASSWORD=your_api_password \
  dataforseo/mcp:latest

Use your API login and password from the API Access section of the dashboard.

3

The MCP server is now available locally at:
http://localhost:3000
Connect any MCP-compatible client to that URL. See the Docker Hub repository for advanced configuration.

1

Point your MCP-compatible client to the official DataForSEO remote endpoint:
https://mcp.dataforseo.com/mcp
2
Authenticate with a Base64-encoded Basic Auth header:
Header
```
Authorization: Basic 
```
Get your credentials from the API Access dashboard. The Send by Email button delivers a pre-encoded token.
3

If your client expects a separate SSE endpoint, use https://mcp.dataforseo.com/http.

1
Make sure Node.js 18+ is installed, then launch the server locally:
Bash
```
npx -y dataforseo-mcp-server
```

Provide your credentials as environment variables:

Bash

export DATAFORSEO_USERNAME=your_api_login
export DATAFORSEO_PASSWORD=your_api_password

Or register the server directly inside your client config:

JSON

{
  "mcpServers": {
    "dataforseo": {
      "command": "npx",
      "args": ["-y", "dataforseo-mcp-server"],
      "env": {
        "DATAFORSEO_USERNAME": "your_api_login",
        "DATAFORSEO_PASSWORD": "your_api_password"
      }
    }
  }
}

Need a different setup? See the full DataForSEO MCP help center.

What's inside

What your assistant can do with it

Seven DataForSEO APIs, exposed as tools any MCP-compatible agent can call.

Keyword research on demand

Search volume, CPC, difficulty, intent, and suggestions — pulled live from Keywords Data and Labs APIs mid-conversation.

SERPs & rankings

Real search results for any query and location, so competitive analysis is grounded in what Google actually shows.

Technical audits

On-Page API checks through natural language: “audit this page and list the top issues” returns real crawl data.

Backlinks & domain intel

Link profiles, competitor comparisons, and domain analytics — formatted into tables and recommendations by your agent.

Start asking with $1 free credit

Who it's for

Who plugs SEO data into their AI

SEO specialists

Research at conversation speed

Keyword clusters, competitor snapshots, and audit summaries in the same chat where you write the strategy.

Content teams

Data-grounded drafts

Brief and draft with real search volumes and SERP context — content built on data instead of vibes.

Builders & automators

Agent workflows that ship

Wire the MCP server into Claude Code, Cursor, or n8n agents and productize SEO deliverables end to end.

SEO software

Prototype before you build

Validate data-driven features in a chat before committing engineering time to a full integration.

Start free — $1 credit included

Pricing

Pay for data, not for seats

The MCP server is free and open source. You pay only standard pay-as-you-go rates for the data your assistant pulls.

Endpoint	Price	Turnaround
MCP server & connection	free — open source on GitHub	npx or hosted HTTP
100 related keywords	≈ $0.024 in API usage	seconds
SERP check (Live)	from $0.002 per query	seconds
Backlink summary	$0.024 + $0.000036 per row	~2 s

$1 free trial credit
$50 minimum top-up
credits never expire
24/7 human support

Try for free Full pricing details

Social proof

Data industry leaders trust DataForSEO

"I always recommend DataForSEO. No worries about server capacity, no need for maintenance, and splendid possibilities for rapid development."

"We found DataForSEO online while planning our beta. We were impressed by the responsive team and comprehensive APIs."

"DataForSEO has always been in our toolkit. When exploring an API, you see what it's truly capable of. Documentation is awesome."

No-code integration with DataForSEO APIs

Easily integrate DataForSEO APIs into your favourite tools without writing a single line of code. With a collection of ready-made connectors and modules you can access high-quality SEO data effortlessly, saving your time and resources.

ChatGPT Make Сursor Claude.ai Postman Gemini Google Sheets Connector Airtable Zapier n8n View Integrations →

Implementation

Release in days, not weeks

Docs that answer first

Every endpoint, parameter, and response field documented with live examples.

Read the docs →

Sandbox & API Explorer

Prototype against the Sandbox for free and fire test calls from the browser before you write a line of code.

Open the Sandbox →

AI assistant on call

Ask it anything about the API — it answers from the docs and drafts integration scripts for you.

Ask the AI assistant →

24/7 human support

Real engineers on the other end — during evaluation, integration, and every day after.

Talk to us →

What is an SEO MCP server?

MCP (Model Context Protocol) is an open standard that lets AI assistants call external tools. The DataForSEO MCP server exposes seven SEO APIs — SERP, Keywords Data, Labs, Backlinks, On-Page, Business Data, and Domain Analytics — as tools that Claude, Cursor, and other MCP clients can use in conversation.

How do I set it up?

Add one entry to your MCP config running “npx -y dataforseo-mcp-server” with your DataForSEO credentials, or connect to the hosted endpoint at mcp.dataforseo.com/mcp. The server is open source at github.com/dataforseo/mcp-server-typescript.

Which AI tools does it work with?

Any MCP-compatible client — including Claude Desktop, Claude Code, Cursor, and n8n’s MCP integration.

Does the MCP server cost extra?

No. The server itself is free; you pay only the standard pay-as-you-go API rates for data your assistant actually requests — for example, about $0.024 for 100 related keywords.

Do I need to know how to code?

No. Setup is a config snippet, and everything after that is plain English. Developers can go deeper, but it’s not required.

How do I get credentials?

Register a DataForSEO account — it takes a minute and comes with $1 in free API credits, enough for plenty of test questions. Credits never expire.

Can the assistant really be trusted with the numbers?

The numbers come from DataForSEO APIs, not the model’s imagination — the same data that powers hundreds of SEO tools. The assistant formats and interprets; the data stays real.

Stop letting your AI guess about SEO

Try for free Read the docs

$1 in Free Credits · No Credit Card · Pay as You Go from $50

Give your AI assistant real SEO data

Your AI is smart. Its SEO data is made up.

Hallucinated metrics

Copy-paste research loops

Automation still needs a developer

Get Started With DataForSEO MCP

Start your unlimited free trial

What your assistant can do with it

Keyword research on demand

SERPs & rankings

Technical audits

Backlinks & domain intel

Who plugs SEO data into their AI

Research at conversation speed

Data-grounded drafts

Agent workflows that ship

Prototype before you build

Pay for data, not for seats

Data industry leaders trust DataForSEO

No-code integration with DataForSEO APIs

Release in days, not weeks

Docs that answer first

Sandbox & API Explorer

AI assistant on call

24/7 human support

Frequently asked questions

What is an SEO MCP server?

How do I set it up?

Which AI tools does it work with?

Does the MCP server cost extra?

Do I need to know how to code?

How do I get credentials?

Can the assistant really be trusted with the numbers?

Stop letting your AI guess about SEO