GeoServer MCP Server

A Model Context Protocol (MCP) server implementation that connects Large Language Models (LLMs) to the GeoServer REST API, enabling AI assistants to interact with geospatial data and services.

Version 0.4.0 (Alpha) is under active development and will be released shortly. We are open to contributions and welcome developers to join us in building this project.

🎥 Demo

📋 Table of Contents

• Features
• Prerequisites
• Installation
  • Docker Installation
  • pip Installation
  • Development Installation
• Available Tools
  • Workspace and Layer Management
  • Data Operations
  • Visualization
• Client Development
  • List Workspaces
  • Get Layer Information
  • Query Features
  • Generate Map
• Planned Features
• Contributing
• License
• Related Projects
• Support
• Badges

🚀 Features

• 🔍 Query and manipulate GeoServer workspaces, layers, and styles
• 🗺️ Execute spatial queries on vector data
• 🎨 Generate map visualizations
• 🌐 Access OGC-compliant web services (WMS, WFS)
• 🛠️ Easy integration with MCP-compatible clients

📋 Prerequisites

• Python 3.10 or higher
• Running GeoServer instance with REST API enabled
• MCP-compatible client (like Claude Desktop or Cursor)
• Internet connection for package installation

🛠️ Installation

Choose the installation method that best suits your needs:

Installing via Smithery

To install GeoServer MCP Server for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @mahdin75/geoserver-mcp --client claude

🛠️ Installation (Docker)

The Docker installation is the quickest and most isolated way to run the GeoServer MCP server. It's ideal for:

• Quick testing and evaluation
• Production deployments
• Environments where you want to avoid Python dependencies
• Consistent deployment across different systems

1. Run geoserver-mcp:

docker pull mahdin75/geoserver-mcp
docker run -d mahdin75/geoserver-mcp

1. Configure the clients:

If you are using Claude Desktop, edit claude_desktop_config.json If you are using Cursor, Create .cursor/mcp.json

{
  "mcpServers": {
    "geoserver-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GEOSERVER_URL=http://localhost:8080/geoserver",
        "-e",
        "GEOSERVER_USER=admin",
        "-e",
        "GEOSERVER_PASSWORD=geoserver",
        "-p",
        "8080:8080",
        "mahdin75/geoserver-mcp"
      ]
    }
  }
}

🛠️ Installation (pip)

The pip installation is recommended for most users who want to run the server directly on their system. This method is best for:

• Regular users who want to run the server locally
• Systems where you have Python 3.10+ installed
• Users who want to customize the server configuration
• Development and testing purposes

1. Install uv package manager.

pip install uv

1. Create the Virtual Environment (Python 3.10+):

Linux/Mac:

uv venv --python=3.10

Windows PowerShell:

uv venv --python=3.10

1. Install the package using pip:

uv pip install geoserver-mcp

1. Configure GeoServer connection:

Linux/Mac:

export GEOSERVER_URL="http://localhost:8080/geoserver"
export GEOSERVER_USER="admin"
export GEOSERVER_PASSWORD="geoserver"

Windows PowerShell:

$env:GEOSERVER_URL="http://localhost:8080/geoserver"
$env:GEOSERVER_USER="admin"
$env:GEOSERVER_PASSWORD="geoserver"

1. Start the server:

If you are going to use Claude desktop you don't need this step. For cursor or your own custom client you should run the following code.

Linux:

source .venv/bin/activate

geoserver-mcp

or

source .venv/bin/activate

geoserver-mcp --url http://localhost:8080/geoserver --user admin --password geoserver --debug

Windows PowerShell:

.\.venv\Scripts\activate
geoserver-mcp

or

.\.venv\Scripts\activate
geoserver-mcp --url http://localhost:8080/geoserver --user admin --password geoserver --debug

1. Configure Clients:

If you are using Claude Desktop, edit claude_desktop_config.json If you are using Cursor, Create .cursor/mcp.json

Windows:

{
  "mcpServers": {
    "geoserver-mcp": {
      "command": "C:\\path\\to\\geoserver-mcp\\.venv\\Scripts\\geoserver-mcp",
      "args": [
        "--url",
        "http://localhost:8080/geoserver",
        "--user",
        "admin",
        "--password",
        "geoserver"
      ]
    }
  }
}

Linux:

{
  "mcpServers": {
    "geoserver-mcp": {
      "command": "/path/to/geoserver-mcp/.venv/bin/geoserver-mcp",
      "args": [
        "--url",
        "http://localhost:8080/geoserver",
        "--user",
        "admin",
        "--password",
        "geoserver"
      ]
    }
  }
}

🛠️ Development installation

The development installation is designed for contributors and developers who want to modify the codebase. This method is suitable for:

• Developers contributing to the project
• Users who need to modify the source code
• Testing new features
• Debugging and development purposes

1. Install uv package manager.

pip install uv

1. Create the Virtual Environment (Python 3.10+):

uv venv --python=3.10

1. Install the package using pip:

uv pip install -e .

1. Configure GeoServer connection:

Linux/Mac:

export GEOSERVER_URL="http://localhost:8080/geoserver"
export GEOSERVER_USER="admin"
export GEOSERVER_PASSWORD="geoserver"

Windows PowerShell:

$env:GEOSERVER_URL="http://localhost:8080/geoserver"
$env:GEOSERVER_USER="admin"
$env:GEOSERVER_PASSWORD="geoserver"

1. Start the server:

If you are going to use Claude desktop you don't need this step. For cursor or your own custom client you should run the following code.

Linux:

source .venv/bin/activate

geoserver-mcp

or

source .venv/bin/activate

geoserver-mcp --url http://localhost:8080/geoserver --user admin --password geoserver --debug

Windows PowerShell:

.\.venv\Scripts\activate
geoserver-mcp

or

.\.venv\Scripts\activate
geoserver-mcp --url http://localhost:8080/geoserver --user admin --password geoserver --debug

1. Configure Clients:

If you are using Claude Desktop, edit claude_desktop_config.json If you are using Cursor, Create .cursor/mcp.json

Windows:

{
  "mcpServers": {
    "geoserver-mcp": {
      "command": "C:\\path\\to\\geoserver-mcp\\.venv\\Scripts\\geoserver-mcp",
      "args": [
        "--url",
        "http://localhost:8080/geoserver",
        "--user",
        "admin",
        "--password",
        "geoserver"
      ]
    }
  }
}

Linux:

{
  "mcpServers": {
    "geoserver-mcp": {
      "command": "/path/to/geoserver-mcp/.venv/bin/geoserver-mcp",
      "args": [
        "--url",
        "http://localhost:8080/geoserver",
        "--user",
        "admin",
        "--password",
        "geoserver"
      ]
    }
  }
}

🛠️ Available Tools

🛠️ Workspace and Layer Management

Tool	Description
list_workspaces	Get available workspaces
create_workspace	Create a new workspace
get_layer_info	Get detailed layer metadata
list_layers	List layers in a workspace
create_layer	Create a new layer
delete_resource	Remove resources

🛠️ Data Operations

Tool	Description
query_features	Execute CQL queries on vector data
update_features	Modify feature attributes
delete_features	Remove features based on criteria

🛠️ Visualization

Tool	Description
generate_map	Create styled map images
create_style	Define new SLD styles
apply_style	Apply existing styles to layers

🛠️ Client Development

If you're planning to develop your own client to interact with the GeoServer MCP server, you can find inspiration in the example client implementation at examples/client.py. This example demonstrates:

• How to establish a connection with the MCP server
• How to send requests and handle responses
• Basic error handling and connection management
• Example usage of various tools and operations

The example client serves as a good starting point for understanding the protocol and implementing your own client applications.

Also, here is the example usgage:

List Workspaces

Tool: list_workspaces
Parameters: {}
Response: ["default", "demo", "topp", "tiger", "sf"]

Get Layer Information

Tool: get_layer_info
Parameters: {
"workspace": "topp",
"layer": "states"
}

Query Features

Tool: query_features
Parameters: {
"workspace": "topp",
"layer": "states",
"filter": "PERSONS > 10000000",
"properties": ["STATE_NAME", "PERSONS"]
}

Generate Map

Tool: generate_map
Parameters: {
"layers": ["topp:states"],
"styles": ["population"],
"bbox": [-124.73, 24.96, -66.97, 49.37],
"width": 800,
"height": 600,
"format": "png"
}

🔮 Planned Features

• Coverage and raster data management
• Security and access control
• Advanced styling capabilities
• WPS processing operations
• GeoWebCache integration

🤝 Contributing

We welcome contributions! Here's how you can help:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Please ensure your PR description clearly describes the problem and solution. Include the relevant issue number if applicable.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Related Projects

• Model Context Protocol - The core MCP implementation
• GeoServer REST API - Official GeoServer REST documentation
• GeoServer REST Python Client - Python client for GeoServer REST API

📞 Support

For support, please Open an issue

🏆 Badges