ThinkGeo MCP Server¶

The ThinkGeo MCP Server allows AI assistants and development tools to access ThinkGeo documentation and resources directly through the Model Context Protocol (MCP).

By connecting an MCP‑compatible AI client to the ThinkGeo Documentation MCP endpoint, developers can ask natural‑language questions about ThinkGeo APIs, libraries, and mapping workflows and receive accurate answers based on the latest ThinkGeo documentation, community posts and samples.

MCP Server URL: https://ai.thinkgeo.com/mcp

What Is MCP?¶

The Model Context Protocol (MCP) is an open standard that allows AI applications to connect to external tools, services, and data sources in real time. Instead of relying solely on their training data, AI models can query MCP servers to retrieve up‑to‑date information or execute specialized tools.

What's Included¶

The server indexes thousands of documents across the full ThinkGeo product ecosystem:

What You Can Do With It¶

Once connected, you can ask your AI assistant natural language questions about ThinkGeo APIs and expect answers grounded in real documentation. Some examples:

• "Show me how to add a ShapeFileFeatureLayer to a WPF map."
• "How do I style features using ClassBreakStyle in WinForms?"
• "Generate a Blazor page that displays a ThinkGeo map with a background tile overlay."
• "Create an ASP.NET Core WebAPI tile service for my shapefile and show it on a Leaflet map."
• "What WFS endpoints does ThinkGeo GIS Server support?"
• "How do I add routing to a .NET MAUI app using ThinkGeo Cloud?"

Or, use Claude Code or Codex directly in your ThinkGeo maps project and ask it for help generating new features and fixing bugs.

Available MCP Tools¶

The server exposes the following tools to MCP-compatible AI clients:

For best results, also install the ThinkGeo AI Agent Skills. They teach your assistant ThinkGeo workflows and common pitfalls, and tell it when to use each of these tools.

Setting Up in Chat Assistants¶

These apps change their menus often, so the steps below may not match exactly. In each one, you're adding a remote MCP server (sometimes called a connector or custom app) with the URL https://ai.thinkgeo.com/mcp. The server doesn't need a sign-in.

Claude (claude.ai and Claude Desktop)¶

1. Go to Customize → Connectors.
2. Click + Add, then Add custom connector.
3. Enter the name thinkgeo-docs and the URL https://ai.thinkgeo.com/mcp. Choose no sign-in if asked.
4. Click Add. In a chat, use the + button to turn on the connector.

On Team and Enterprise plans, an owner first adds the connector in Organization settings → Connectors, and then each member connects it. One connector works in claude.ai, Claude Desktop, and the Claude mobile apps. See Anthropic's guide to custom connectors.

ChatGPT¶

Custom MCP servers are available on some paid ChatGPT plans, and Business and Enterprise workspaces may need an admin to allow them.

1. Go to chatgpt.com/plugins, select +, then Add custom MCP server. (On some plans this is under Settings → Apps → Advanced settings, with developer mode turned on.)
2. Enter the name thinkgeo-docs and the URL https://ai.thinkgeo.com/mcp, and choose no authentication.
3. Accept the warning about custom servers and create it.

See OpenAI's connect to ChatGPT guide for the current steps.

Gemini¶

In the Gemini web app, some accounts can add an MCP server as a custom app: go to Settings → Connected Apps, click Add a custom app under Custom apps, and enter https://ai.thinkgeo.com/mcp. This feature is limited by plan, account type, and country; see Google's help page for who can use it. For the Gemini CLI, see below.

Grok¶

1. Go to grok.com/connectors.
2. Select New Connector, then Custom.
3. Enter https://ai.thinkgeo.com/mcp and name it thinkgeo-docs.

See xAI's connectors documentation.

Setting Up in Developer Tools and IDEs¶

Claude Code¶

Run:

claude mcp add --transport http thinkgeo-docs https://ai.thinkgeo.com/mcp

Add --scope project to share it with your team through a .mcp.json file in your project, or --scope user to use it in all your projects. You can also create the .mcp.json file in your project root yourself:

{
  "mcpServers": {
    "thinkgeo-docs": {
      "type": "http",
      "url": "https://ai.thinkgeo.com/mcp"
    }
  }
}

Codex¶

Run:

codex mcp add thinkgeo-docs --url https://ai.thinkgeo.com/mcp

Visual Studio Code (GitHub Copilot)¶

1. Open the Command Palette (Ctrl+Shift+P) and run MCP: Add Server.
2. Choose HTTP, enter https://ai.thinkgeo.com/mcp, and name it thinkgeo-docs.

Or add it to a configuration file yourself: .vscode/mcp.json in your workspace, or your user configuration (run MCP: Open User Configuration):

{
  "servers": {
    "thinkgeo-docs": {
      "type": "http",
      "url": "https://ai.thinkgeo.com/mcp"
    }
  }
}

Then open Copilot Chat, switch to Agent mode, and make sure the ThinkGeo tools are turned on in the Tools picker. Older VS Code versions kept MCP servers under an "mcp" key in settings.json; current versions move them to mcp.json automatically.

Visual Studio¶

Requires Visual Studio 2022 version 17.14 or later, or Visual Studio 2026, with GitHub Copilot.

1. In the Copilot Chat window, switch to Agent mode.
2. Select Tools, then the + button. (In Visual Studio 2026, choose Add custom MCP server.)
3. Enter the name thinkgeo-docs and the URL https://ai.thinkgeo.com/mcp.
4. Turn on the ThinkGeo tools in the Tools list.

Or create a .mcp.json file in your solution folder (to share it through source control) or in %USERPROFILE% (for all solutions):

{
  "servers": {
    "thinkgeo-docs": {
      "url": "https://ai.thinkgeo.com/mcp"
    }
  }
}

See Microsoft's guide to MCP servers in Visual Studio.

Cursor¶

Add the server to .cursor/mcp.json in your project, or ~/.cursor/mcp.json for all projects:

{
  "mcpServers": {
    "thinkgeo-docs": {
      "url": "https://ai.thinkgeo.com/mcp"
    }
  }
}

The ThinkGeo tools are then available to Cursor's agent. See Cursor's MCP documentation.

JetBrains Rider and IntelliJ IDEA¶

AI Assistant:

1. Open Settings → Tools → AI Assistant → Model Context Protocol (MCP).

2. Click Add, choose HTTP, and enter:

   {
     "mcpServers": {
       "thinkgeo-docs": {
         "url": "https://ai.thinkgeo.com/mcp"
       }
     }
   }
   

3. Click Apply.

Junie: add the same JSON to .junie/mcp/mcp.json in your project, or ~/.junie/mcp/mcp.json for all projects.

See JetBrains' AI Assistant MCP documentation.

Gemini CLI¶

Run:

gemini mcp add --transport http thinkgeo-docs https://ai.thinkgeo.com/mcp

This adds the server to the current project. Add -s user to use it in all your projects. Or add it to .gemini/settings.json in your project, or ~/.gemini/settings.json:

{
  "mcpServers": {
    "thinkgeo-docs": {
      "httpUrl": "https://ai.thinkgeo.com/mcp"
    }
  }
}

Tips for Getting the Best Results¶

• Be specific about the product. Mention the platform you're targeting — WPF, WinForms, Blazor, MAUI, WebAPI, or GIS Server — so the AI can search the right namespace.
• Install the Agent Skills. The ThinkGeo AI Agent Skills add workflows and known pitfalls on top of the documentation, so generated code is far more likely to compile.
• Ask for code examples. The server includes full "How Do I" solution files, so asking for working code samples will often yield complete, runnable snippets.
• Combine with your project context. Paste in a relevant class or method from your project alongside your ThinkGeo question for the most accurate AI-generated code.
• Always review generated code. AI output can vary depending on your prompt and model. Verify generated code for correctness, architectural fit, and security before using it in production.

Need Help?¶

• Documentation: docs.thinkgeo.com
• Community Forum: community.thinkgeo.com
• Support Helpdesk: helpdesk.thinkgeo.com