Configure Blender MCP Using Gemini CLI

Introduction

Gemini CLI is a powerful command-line interface that allows you to interact with Google's Gemini AI model directly from your terminal. By integrating Blender MCP with Gemini CLI, you can leverage AI capabilities to enhance your 3D modeling workflow in Blender.

This guide will walk you through the entire setup process, from installing Node.js to verifying that your MCP service is working correctly with Gemini CLI.

Prerequisites

Windows, macOS, or Linux system
Administrator access to install software
A Google account with Gemini access
Basic familiarity with command-line interfaces
Blender installed on your system (optional, but recommended)

Step 1: Install Node.js

Gemini CLI and the Blender MCP server require Node.js to run. Follow these steps to install it:

For Windows:

Visit nodejs.org
Download the LTS (Long-Term Support) version
Run the installer and follow the installation wizard
Accept all default options (this includes npm and PATH configuration)
Click "Finish" to complete the installation

For macOS:

Using Homebrew: brew install node
Or download from nodejs.org and run the installer

For Linux:

Use your package manager. For example, on Ubuntu/Debian:

sudo apt update
sudo apt install nodejs npm

Verify Node.js Installation

Open a terminal or command prompt and run:

node --version
npm --version

Both commands should return version numbers if Node.js is correctly installed.

Step 2: Download and Install Gemini CLI

Now that Node.js is installed, you can download and set up Gemini CLI:

Installation Steps

Open a terminal or command prompt
Navigate to or create a directory where you want to store your Gemini CLI project (we'll call it gmcp)
Create the directory:
```
mkdir gmcp
cd gmcp
```

Download Gemini CLI using npm:

npm install -g @google/generative-ai-cli

Alternatively, use npx to run it directly without global installation:
```
npx @google/generative-ai-cli
```

Step 3: Configure Blender MCP Service

The MCP (Model Context Protocol) service configuration is the key step that connects Blender MCP with Gemini CLI.

Locating the Configuration File

The Gemini CLI stores its configuration in a settings file located at:

D:\seocode\tmp\gmcp\.gemini\settings.json

Adding the Configuration

Open the settings.json file with your preferred text editor (e.g., Visual Studio Code, Notepad++, or even Notepad).

Add or update the following configuration to enable the Blender MCP service:

{
  "mcpServers": {
    "blender": {
      "command": "uvx",
      "args": [
        "blender-mcp"
      ]
    }
  }
}

Configuration Explanation

mcpServers: An object containing all MCP service definitions
blender: The name of your MCP service (used to identify it in Gemini CLI)
command: The command to execute (uvx is a package runner for Python)
args: Arguments passed to the command (blender-mcp is the Blender MCP package)

Save the Configuration

After adding the configuration, save the file. Make sure the JSON is properly formatted with no trailing commas or syntax errors.

Step 4: Exit and Log Back Into Gemini CLI

To apply the new MCP service configuration, you need to restart Gemini CLI:

Exit Gemini CLI

If Gemini CLI is running, exit it by typing:

exit

Log Back In

Restart Gemini CLI in your terminal:

npx @google/generative-ai-cli

Or if you installed it globally:

gemini-cli

Step 5: Verify MCP Service Configuration

Once you've logged back into Gemini CLI, verify that the Blender MCP service is properly configured and accessible:

Check Service Status

In Gemini CLI, you should see output indicating that the MCP service has been loaded. Look for messages similar to:

[INFO] Loading MCP services...
[INFO] Blender MCP service loaded successfully
[INFO] Available tools: blender_create_object, blender_modify_object, ...

Test the Integration

Try issuing a simple command that uses the Blender MCP service. For example:

Create a cube in Blender

If Blender is running with the MCP plugin installed, you should see the cube appear in your Blender scene. Gemini will process your request and send commands through the MCP protocol to Blender.

Troubleshooting

Issue: Settings.json not found

Solution: Manually create the .gemini folder in your project directory and create a settings.json file with the configuration provided above.

Issue: "uvx command not found"

Solution: Install the required Python tools package:

pip install uv

Issue: MCP service not loading

Solution: Check the following:

Verify JSON syntax in settings.json is correct (use a JSON validator if needed)
Ensure the file path is correct: D:\seocode\tmp\gmcp\.gemini\settings.json
Check that Blender MCP is installed: uvx blender-mcp --version
Review Gemini CLI logs for detailed error messages

Issue: Blender not responding to commands

Solution: Ensure that:

Blender is running with the MCP plugin installed
The MCP plugin is enabled in Blender settings
Both Gemini CLI and Blender are connected on the correct network port

Next Steps

Now that you have Blender MCP configured with Gemini CLI, you can:

Create complex 3D scenes using natural language prompts
Modify objects and materials with AI assistance
Automate repetitive 3D modeling tasks
Combine Gemini's AI capabilities with Blender's powerful 3D tools
Explore advanced MCP features for custom scripting and automation

For more information on using Blender MCP, visit the official GitHub repository.

Conclusion

Setting up Blender MCP with Gemini CLI opens up a world of possibilities for AI-assisted 3D modeling. While the configuration process involves several steps, following this guide should have you up and running in no time. Remember that the key to success is ensuring your configuration files are properly formatted and all components are correctly installed.

If you encounter any issues or have questions, refer to the troubleshooting section or consult the Blender MCP documentation.