mcp

In the era of AI-driven search (Generative Engine Optimization or GEO), the speed and accessibility of data are critical. Imagine being able to query your brand's AI visibility data directly within your conversation with an AI assistant—whether in Claude Desktop or Cursor IDE—without ever logging into a dashboard.

Geoly MCP (Model Context Protocol) makes this possible.

This guide dives deep into Geoly's MCP capabilities, practical use cases, and step-by-step configuration for leading AI editors, empowering you to build your own personal AI SEO analyst.

What is MCP (Model Context Protocol)?

The Model Context Protocol (MCP) is a standardized interface that allows AI assistants (like Claude, ChatGPT, etc.) to connect directly to local or remote data sources.

By integrating Geoly's MCP service, your AI assistant evolves from a simple chatbot into a powerful tool with direct access to Geoly's real-time database. You can ask questions in natural language, and the AI will automatically invoke the appropriate tools to fetch data, providing deep analysis in context—eliminating the need for manual report checking.

The Geoly MCP Toolkit

Geoly currently offers 10 powerful MCP tools, covering everything from high-level brand monitoring to granular prompt analysis.

1. Overview Tools (4)

Designed for a quick pulse check on brand performance.

2. Deep Dive Tools (6)

Built for in-depth investigation and optimization opportunities.

Common Use Cases

Once connected, you can perform complex analyses through simple conversation. Here are three high-impact scenarios:

Scenario 1: Detecting Lost Referrals

Problem: Traffic has dropped suddenly, and you suspect a key AI citation has been lost. Prompt: "Check the citation trend for blog.example.com over the last 30 days. Are there any sudden drops?" AI Action: Invokes the Domain Details tool to retrieve daily citation trend data, allowing you to visualize exactly when and where a referral source might have disappeared.

Scenario 2: Uncovering Content Gaps

Problem: You want to identify high-value prompts where your content is invisible. Prompt: "Which monitored prompts are not citing example.com?" AI Action: Uses the Content Gap Analysis tool to compare your monitored list against actual citations, generating a list of missing opportunities. These are your immediate targets for content optimization.

Scenario 3: Competitor Analysis

Problem: You need to understand how AI answers a core keyword using competitor data. Prompt: "How does the prompt 'best CRM software' perform on ChatGPT vs. Gemini? Are competitors mentioned?" AI Action: Calls Prompt Details to return visibility scores, rankings, and specific competitor mentions for that prompt across platforms.

How to Configure Geoly MCP

Configuration is straightforward. You only need two pieces of information to set up your AI client.

Prerequisites

1. Log in to the Geoly Dashboard.
2. Navigate to the MCP Menu.
3. Create and copy the following:
   • MCP Server URL: e.g., https://app.geoly.ai/api/mcp
   • API Token: e.g., geom_xxxxxxxxxxxx

Integration Guides

1. Claude Desktop (Recommended)

Currently offers the best MCP experience.

1. Open Claude Desktop.
2. Go to Settings → Developer → Edit Config.
3. Add the geoly server configuration to the JSON file:

{
  "mcpServers": {
    "geoly": {
      "url": "https://app.geoly.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer geom_YOUR_TOKEN"
      }
    }
  }
}

4. Save and restart Claude Desktop.
5. Look for the tool icon (hammer) in the chat interface to confirm Geoly tools are loaded.

2. Cursor

Ideal for developers who want SEO data alongside their code.

1. Open Cursor → Settings → MCP.
2. Click "Add new MCP server".
3. Fill in the details:
   • Type: streamable-http
   • Name: geoly
   • URL: https://app.geoly.ai/api/mcp
4. After saving, open ~/.cursor/mcp.json and manually add the authentication header:

{
  "mcpServers": {
    "geoly": {
      "url": "https://app.geoly.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer geom_YOUR_TOKEN"
      }
    }
  }
}

5. Restart Cursor and use it in Agent mode.

3. Windsurf

1. Open Windsurf → Settings → MCP Configuration.
2. Click "Add custom server" and edit the config:

{
  "mcpServers": {
    "geoly": {
      "serverUrl": "https://app.geoly.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer geom_YOUR_TOKEN"
      }
    }
  }
}

3. Save and restart Windsurf.

4. Claude Code (CLI)

For terminal users, quickly add Geoly via Claude Code:

claude mcp add geoly --transport streamable-http https://app.geoly.ai/api/mcp \
  --header "Authorization: Bearer geom_YOUR_TOKEN"

5. Antigravity

You can add Geoly MCP in Antigravity by editing the raw MCP config.

1. Open MCP Servers and select Manage MCP Servers.
2. Click View raw config.
3. Paste the following JSON and save:

{
  "mcpServers": {
    "geoly": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://app.geoly.ai/api/mcp",
        "--header",
        "Authorization:Bearer geom_YOUR_TOKEN"
      ]
    }
  }
}

4. Refresh the MCP list and confirm Geoly tools are available.

6. Trae IDE

You can add Geoly MCP in Trae via manual configuration.

1. Open Trae and go to Settings.
2. Navigate to MCP, then click Add -> Manual.
3. Paste the following JSON and confirm:

{
  "mcpServers": {
    "geoly": {
      "url": "https://app.geoly.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer geom_YOUR_TOKEN"
      }
    }
  }
}

4. Verify Geoly MCP appears in the MCP list and is enabled.

Pro Tips

• Natural Language: No need to memorize tool names. Just describe what you want (e.g., "Analyze my brand visibility"), and the AI will select the right tool.
• Filtering: If data is overwhelming, add constraints like "Show only the top 5 results," "Focus on ChatGPT," or "Last 90 days."
• Multi-language: Geoly MCP supports queries in both English and Chinese, returning results in your preferred language.

Start your AI-native SEO journey today by integrating Geoly MCP into your daily workflow!