- Explore MCP Servers
- Spline-MCP
Spline Mcp
What is Spline Mcp
Spline-MCP is an MCP server designed to facilitate interaction between AI models and the Spline 3D design tool API. It serves as a bridge, allowing AI assistants to manage and create 3D designs through simple commands.
Use cases
Use cases for Spline-MCP include automating the creation of 3D models, generating animations based on user interactions, and enabling AI assistants to assist non-technical users in designing 3D scenes.
How to use
To use Spline-MCP, install it via npm or from the source. Ensure you have Node.js and a Spline account with an API key. After installation, configure the server by creating a .env file with your API key and start using it to interact with Spline.
Key features
Key features of Spline-MCP include exporting scenes in various formats (GLB, GLTF, FBX, OBJ), importing 3D models, creating keyframe animations, triggering animations, and managing Spline projects through a standardized interface.
Where to use
Spline-MCP can be used in various fields including game development, animation production, virtual reality, and any domain requiring 3D design and modeling.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Spline Mcp
Spline-MCP is an MCP server designed to facilitate interaction between AI models and the Spline 3D design tool API. It serves as a bridge, allowing AI assistants to manage and create 3D designs through simple commands.
Use cases
Use cases for Spline-MCP include automating the creation of 3D models, generating animations based on user interactions, and enabling AI assistants to assist non-technical users in designing 3D scenes.
How to use
To use Spline-MCP, install it via npm or from the source. Ensure you have Node.js and a Spline account with an API key. After installation, configure the server by creating a .env file with your API key and start using it to interact with Spline.
Key features
Key features of Spline-MCP include exporting scenes in various formats (GLB, GLTF, FBX, OBJ), importing 3D models, creating keyframe animations, triggering animations, and managing Spline projects through a standardized interface.
Where to use
Spline-MCP can be used in various fields including game development, animation production, virtual reality, and any domain requiring 3D design and modeling.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
Spline MCP Server
An MCP server for working with the Spline 3D design tool API. This server provides a convenient interface for AI models like Claude, GPT, and others to interact with Spline’s features, including exporting scenes, importing models, creating animations, and managing projects.
What is this?
This project allows AI assistants to control Spline 3D design software. Think of it as a bridge between AI and the 3D design world.
For non-technical users: This tool helps your AI assistant work with Spline, allowing it to help you create and manage 3D designs through simple conversations.
For developers: This is an MCP (Model Control Protocol) server implementation for Spline, making the Spline API accessible to AI assistants through a standardized interface.
Features
Basic Operations
- Export Spline scenes to various formats (GLB, GLTF, FBX, OBJ)
- Import 3D models into Spline
- Get details about Spline scenes
- List available Spline scenes
Animation Capabilities
- Create keyframe animations for objects
- Trigger existing animations
- List animations in a scene
- Create event-triggered animations (onClick, onHover, etc.)
Installation
Prerequisites
Before you begin, make sure you have:
- Node.js installed (version 14 or higher)
- A Spline account and API key from Spline
- Basic familiarity with using the command line/terminal
From npm (once published)
# Install the package globally
npm install -g spline-mcp-server
From Source (For Beginners)
Follow these step-by-step instructions:
- Open your terminal/command prompt
- Clone this repository:
git clone https://github.com/Youbadawy/Spline-MCP.git - Navigate to the project folder:
cd Spline-MCP - Install the dependencies:
npm install
Configuration
- Create a file named
.envin the project folder - Add your Spline API credentials to the file:
SPLINE_API_KEY=your_api_key_here SPLINE_API_URL=https://api.spline.design
To get your Spline API key:
- Log in to your Spline account
- Go to your account settings
- Look for the API section to generate a key
Running the Server
Using the CLI (if installed globally)
spline-mcp-server
Using npm scripts (if installed from source)
npm start
You should see a message like:
spline-mcp-server is running on port 3000 Server information available at http://localhost:3000/ API endpoints available at http://localhost:3000/api/* OpenAPI specification available at http://localhost:3000/openapi.json
Testing Your Installation
To verify everything is working:
- Open your web browser
- Go to http://localhost:3000/
- You should see basic server information in JSON format
- Visit http://localhost:3000/openapi.json to view the full API specification
Connecting to AI-Powered IDEs
You can connect this MCP server to various AI-powered IDEs to enhance their capabilities with Spline integration.
Cursor AI
To connect to Cursor AI:
- Make sure the Spline MCP server is running locally
- Create an
mcp_config.jsonfile in your project directory with:(A more detailed version of this file is already included in this repository){ "name": "spline", "url": "http://localhost:3000" } - In Cursor, open the command palette (Ctrl+Shift+P or Cmd+Shift+P)
- Select “Connect to MCP Server”
- Choose “Local” and select your
mcp_config.jsonfile - You can now interact with Spline through Cursor AI by asking questions like “List my available Spline scenes” or “Export my Spline scene to GLB format”
Claude
To connect to Claude (in web interface):
- Make sure the Spline MCP server is running locally
- Share the OpenAPI spec with Claude by providing this URL:
http://localhost:3000/openapi.json - Tell Claude about the API capabilities and how to interact with Spline
- You can now ask Claude to perform Spline operations for you
Cline
To connect to Cline (command-line Claude interface):
- Make sure the Spline MCP server is running locally
- Add the Spline MCP server to your Cline configuration:
cline config set mcp.spline.url "http://localhost:3000" - Enable the MCP server in your Cline session:
cline --enable-mcp=spline - You can now interact with Spline through Cline
Windsurf
To connect to Windsurf:
- Make sure the Spline MCP server is running locally
- In Windsurf settings, go to “Integrations” > “MCP Servers”
- Add a new MCP server with:
- Name: Spline
- URL: http://localhost:3000
- Save and enable the integration
- You can now use Spline features directly in Windsurf
Troubleshooting
If you encounter issues:
-
Server won’t start
- Ensure you have Node.js v14 or higher installed (
node -v) - Check that all dependencies are installed (
npm install) - Verify your .env file is in the correct location and formatted properly
- Ensure you have Node.js v14 or higher installed (
-
API connection issues
- Verify your Spline API key is correct
- Check that the server is running (http://localhost:3000 should be accessible)
- Look for error messages in the terminal where the server is running
-
IDE connection issues
- Verify the server is running locally
- Check that the MCP configuration has the correct URL
- Restart your IDE after connecting to the MCP server
Using the Client
The server includes a JavaScript client that developers can use to interact with it programmatically:
import { McpClient } from 'spline-mcp-server/src/utils/mcpClient.js';
const client = new McpClient({
baseUrl: 'http://localhost:3000'
});
// List scenes
const scenes = await client.listScenes({ limit: 5 });
console.log(scenes);
// Export a scene
const result = await client.exportScene({
sceneId: 'your-scene-id',
format: 'glb'
});
console.log(result);
See the examples directory for more examples.
API Endpoints
The server exposes the following API endpoints:
Basic Operations
POST /api/exportScene: Export a Spline scenePOST /api/importScene: Import a 3D model into SplinePOST /api/getSceneDetails: Get details about a Spline scenePOST /api/listScenes: List available Spline scenes
Animation Operations
POST /api/createAnimation: Create a new animationPOST /api/triggerAnimation: Trigger an existing animationPOST /api/listAnimations: List animations in a scenePOST /api/createEventAnimation: Create an event-triggered animation
OpenAPI Specification
The server provides an OpenAPI specification at /openapi.json that describes all available endpoints and their parameters.
Using with AI Models
This MCP server can be used with AI models to interact with Spline’s features. Here are some examples of what you can ask the AI to do:
Basic Operations
List my available Spline scenes
Export my Spline scene with ID "abc123" to GLB format
Import the 3D model from "https://example.com/model.glb" into my Spline project
Get details for my Spline scene with ID "abc123"
Animation Operations
Create an animation named "Rotate" for the cube object in my scene
Trigger the "Bounce" animation for the ball object
Create an onClick animation that makes an object move up when clicked
List all animations in my scene
Animation Examples
Creating a Simple Rotation Animation
Create an animation named "Spin" for object "cube-123" in scene "abc456" with keyframes for rotation
Creating an Event-Based Animation
Create an onClick animation named "Grow" for object "button-123" in scene "abc456" that scales the object to 1.5x its size
Triggering Animations
Trigger the "Pulse" animation for object "heart-123" in scene "abc456" with loop enabled
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Thanks to @aydinfer for starting the project
https://github.com/aydinfer/spline-mcp-server/tree/main/src
License
MIT
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.










