Skip to main content

πŸ§ͺ [EXPERIMENTAL] Model Context Protocol (MCP) Server

Use Search Atlas AI-powered services in your AI tools.

Updated over a week ago

Search Atlas MCP server overview

The Search Atlas MCP server provides AI agents with secure access to Search Atlas's powerful SEO automation and content creation services.

Specifically, the following Search Atlas services are available via Search Atlas MCP:

  • OTTO SEO - Complete SEO automation, including project management, issue detection, AI-powered recommendations, Complete Site Audit reporting, and automated fix deployment for:

    • Onpage Issues

    • Link Building

    • Indexing

    • Miscellaneous

  • Content Genius (Coming Soon) - Advanced content creation and project management

  • Site Explorer (Coming Soon) - Website analysis and competitive research

  • Local SEO (Coming Soon) - Local business optimization and management

The Search Atlas MCP server lets you integrate Search Atlas services into AI tools such as:

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

πŸ“š Use cases

With Search Atlas MCP, you can access powerful SEO automation through your AI agents and speed up your SEO workflows. For example:

  • New Site Auditing: "I want to start a new project for this example.com website. Can you audit it for me?"

  • Project Management: "List me my top projects" - displays a ranked list with project status, audit metrics, and OTTO installation status

  • Project Analysis: "Give me a summary of this project: mydomain.com" - provides a comprehensive overview with key metrics and SEO health score

  • Targeted Optimization: "Generate title tag suggestions for my top 5 priority pages" - analyzes priority pages and creates optimized suggestions

  • Bulk Deployment: "Deploy all title tag suggestions in bulk" - processes all changes and confirms successful implementation

No more manual SEO work. Just one connection that brings Search Atlas automation straight into your AI workflows.

Create a new project by saying:

  • "I want to start a new project for [domain]. Can you audit it for me?"

  • β€œI want to audit this website: [domain]”

Track your website's health by asking OTTO:

  • "Give me a summary for this project: [project-domain]."

  • "List all top priority fixes for [project-domain]."

You can also ask OTTO to suggest new fixes to improve your website's score:

  • Generate title tag suggestions for my top 5 priority pages on [project-domain].

  • Deploy all title tag suggestions in a single bulk update.

πŸ’¬ Example AI queries

Once you connect an AI tool like Claude to Search Atlas MCP, you can write prompts such as:

"Find my project for nike.com and show me all the SEO issues that need attention."

Claude would connect to the Search Atlas services via Search Atlas MCP, locate your project, fetch issue data, and respond with a comprehensive analysis of SEO problems and recommendations.

"Generate AI recommendations for meta description issues on example.com and then deploy them to the live website."

Your AI agent would automatically generate optimized meta descriptions using Search Atlas AI and deploy them directly to your website through Search Atlas's automation system.

You can also ask for multiple fixes to test our system, if you don't mind. The prompt can be something like:

"Generate title tag suggestions for my top 5 priority pages."

This way, the agent will filter which pages you can get the most results from and deploy them for you.

πŸ”‘ Get Search Atlas MCP access

Access to the Search Atlas MCP server is available to all users with an active Search Atlas subscription.

To get your API key:

  1. From the Search Atlas homepage, go to the Settings icon on the top right

  2. Navigate to the "API Key" tab

  3. Click "Copy API Key Button" to get your unique API key

For detailed instructions, refer to the Search Atlas Settings Knowledge Base.

πŸ“ˆ API usage and quotas

The Search Atlas MCP server connects to Search Atlas products via APIs. We don't limit API calls from MCP to our Product APIs, but the underlying products contain tools that are limited by quota.

Tool usage quotas are tied to your existing Search Atlas subscription plan. There are no additional charges for MCP utilization - you can use Search Atlas MCP within your current plan's tool limits.

Please feel free to contact support if you have questions about your plan's tool quotas.

πŸ€– AI-generated responses disclaimer

The Search Atlas MCP brings Search Atlas services into your AI agents for SEO automation and content creation. AI responses may be incomplete or inaccurate; therefore, always verify SEO changes before deploying them to production websites. Search Atlas cannot guarantee the accuracy or reliability of AI agents and is not liable for any outcomes resulting from their use.

βš™οΈ Set up Search Atlas MCP server

AI agent requirements

Make sure your AI agent:

  • Supports HTTP transport

  • Can send custom headers for API key authentication

  • Supports JSON-RPC 2.0 protocol

Search Atlas MCP Server Endpoint

This server uses HTTP transport. The Search Atlas MCP endpoint is:

Authentication is handled through API key headers:

X-API-KEY: your_searchatlas_api_key_here

🌐 Search Atlas MCP Server – Community

The MCP Server Community edition is a simplified and restricted version of the Search Atlas MCP server. It is designed for testing and lightweight usage without requiring authentication.

Key differences from the standard MCP server:

  • No authentication required – you do not need an API key.

  • Throttled usage – strict rate limits are enforced to prevent abuse.

  • Limited availability – intended for evaluation and low-volume testing only.

  • Separate endpoint – all requests must use the /api/v1/mcp/community endpoint.

Community Endpoint:

Use the following endpoint instead of the standard one:

Limitations:

  • Requests may be delayed or blocked if rate limits are exceeded.

  • Not all MCP tools are guaranteed to be available.

  • The Community Edition is not recommended for use in production environments.

For full access, higher quotas, and guaranteed reliability, use the authenticated MCP server with your API key.

🧩 Client-specific setup

Different MCP clients require different configuration formats:

Claude (browser and desktop)

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

  1. Claude (browser and desktop)

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

    1. Navigate to Settings β†’ Connectors β†’ Add custom connector

    2. Name the connector and add the Search Atlas MCP server URL: https://mcp.searchatlas.com/api/v1/mcp

    3. Configure API key authentication as required by your Claude client

    4. Save the changes

    For more information, refer to the official Anthropic documentation.

Claude Code SDK

  1. Navigate to your terminal and paste the following (remember to paste the correct API key).

claude mcp add --transport http searchatlas https://mcp.searchatlas.com/api/v1/mcp --header "X-API-KEY: YOUR_SEARCHATLAS_API_KEY_HERE”

2. Run Claude in your terminal.

Claude Code Desktop

In Settings β†’ Developer β†’ Edit Config β†’ claude_desktop_config.json, add the Search Atlas MCP configuration to your Cursor MCP settings:

{

"mcpServers":{

"searchatlas":{

"command":"npx",

"args":[

"mcp-remote",

"--header",

"X-API-KEY: ${X_API_KEY}"

],

"env":{

"X_API_KEY":"YOUR_SEARCHATLAS_API_KEY_HERE"

}

}

}

}

}

Cursor

​​ Loom

In Settings β†’ Cursor Settings β†’ MCP & Integrations, add the Search Atlas MCP configuration to your Cursor MCP settings:

Search Atlas MCP configuration (JSON)

{

"mcpServers": {

"searchatlas": {

"headers": {

"X-API-KEY": "YOUR_SEARCHATLAS_API_KEY_HERE"

}

}

}

}

Replace YOUR_SEARCHATLAS_API_KEY_HERE with your actual API key from Search Atlas Settings.

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

VS Code

To add the Search Atlas 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:

Search Atlas MCP configuration (JSON)

{

"mcpServers": {

"searchatlas": {

"headers": {

"X-API-KEY": "YOUR_SEARCHATLAS_API_KEY_HERE"

}

}

}

}

Replace YOUR_SEARCHATLAS_API_KEY_HERE with your actual API key from Search Atlas Settings.

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

Warp

In Settings β†’ Settings β†’ AI β†’ Manage MCP Servers β†’ Add, add the Search Atlas MCP configuration to your Cursor MCP settings:

{

"searchatlas": {

"command": "npx",

"args": [

"mcp-remote",

"--header",

"X-API-KEY: ${X_API_KEY}"

],

"env": {

"X_API_KEY": "YOUR_SEARCHATLAS_API_KEY_HERE"

},

"working_directory": null

}

}

Direct HTTP Integration

You can also integrate directly using any HTTP client:

# List available services from Search Atlas MCP

-H "X-API-Key: YOUR_SEARCHATLAS_API_KEY"

# Call a Search Atlas MCP service

-H "Content-Type: application/json" \

-H "X-API-Key: YOUR_SEARCHATLAS_API_KEY" \

-d '{

"jsonrpc": "2.0",

"method": "tools/call",

"params": {

"name": "find_project_by_hostname",

"arguments": {"hostname": "example.com"}

},

"id": 1

}'

🧠 Troubleshooting

Authentication problems

If you're getting authentication errors:

  • Verify your API key is correct and copied from Search Atlas Settings

  • Ensure you have a Pro or Enterprise subscription for API access

  • Check that the X-API-Key header is properly formatted

Connection issues

Issues querying data

  • Confirm you have access to the products you're trying to use with your Search Atlas subscription

  • Check that you haven't exceeded any rate limits for your Search Atlas products

  • Verify your API key has the necessary permissions

Missing services

Your AI agent should be able to access Search Atlas products through the Search Atlas MCP server. If services are not available:

  • Make sure authentication is completed successfully

  • Verify your Search Atlas subscription includes the services you're trying to access

  • Check for any error messages in your AI agent's console

  • Try calling the Search Atlas MCP health check endpoint:

Common error codes

Code

Message

Solution

401

Unauthorized

Check your API key

403

Forbidden

Verify account permissions and subscription

429

Too Many Requests

Implement rate limiting, wait before retrying

-32601

Method not found

Check service name spelling

-32602

Invalid params

Verify required parameters

πŸ’¬ Support

For additional help:

βœ… Closing Note

The Search Atlas MCP Server connects the power of Search Atlas SEO automation directly to your AI tools β€” enabling seamless communication, real-time optimization, and instant content deployment within your preferred environment.

By integrating with the MCP framework, you can:

  • Automate audits, fixes, and SEO recommendations using OTTO SEO

  • Scale content creation and publishing with Content Genius

  • Perform data-driven research through Site Explorer

  • Manage and monitor projects efficiently across multiple clients or brands

This integration bridges AI and SEO workflows, helping you build smarter, faster, and more effective digital strategies.

πŸ’‘ Pro Tip: Combine Search Atlas MCP automation with standard Search Atlas dashboards to maintain complete visibility into your data, reports, and live site performance.

If you encounter issues or need guidance with setup, please take a look at the Troubleshooting and Support sections in this guide, or reach out to our team at any time.

Did this answer your question?