Unity-AI-ParticleSystem

This MCP server provides AI-powered tooling to inspect and modify Unity ParticleSystem components directly from natural language commands. It exposes two primary actions: particle-system-get to fetch the current state of a ParticleSystem (including modules, renderer settings, and state data) and particle-system-modify to update any module properties (such as emission rate, shape, velocity over lifetime, color gradients, and more). By running the server in a container or environment supported by MCP, you can connect your MCP client to query a ParticleSystem in your Unity project and apply changes programmatically, enabling rapid experimentation and procedural effects generation.

To use it, connect your MCP client to the Unity AI Particle System server and issue commands like particle-system-get for a specific GameObject containing a ParticleSystem, or particle-system-modify with a JSON payload describing the desired module changes. The server supports all 24 ParticleSystem modules listed in the documentation (e.g., Main, Emission, Shape, Velocity Over Lifetime, Color Over Lifetime, Trails, Renderer, and Custom Data). This enables you to fine-tune lifetimes, speeds, sizes, colors, texture sheets, collision and external forces, and more, all without manually navigating Unity’s Inspector.

If you already have a Unity project prepared, you can integrate the MCP workflow by connecting to the server from your preferred MCP client, selecting the target ParticleSystem GameObject in your scene, and executing the appropriate get/modify operations. This is particularly useful for rapid prototyping, procedural effect generation, and automating repetitive particle-tuning tasks across multiple assets.

Prerequisites:

• Docker installed and running on your machine (if you choose to use the Docker deployment method).
• Access to the Unity project you want to manage with MCP (the container can be connected to your project files or run in an environment with Unity runtime support).

Option A: Run with Docker

1. Ensure Docker is installed: follow https://docs.docker.com/get-started/
2. Pull and run the Unity AI Particle System MCP server image: docker run -it --rm ivanmurzak/unity-ai-particlesystem
3. Once running, connect your MCP client to the server using the configured endpoint (default local/container interface).

Option B: Customize or build from source

1. Clone the repository hosting the MCP server implementation (if you have access).
2. Follow the repository’s build instructions (requirements.txt or Unity build steps) to create a runnable server binary or container image.
3. Run the server locally or in your CI/CD environment and connect with your MCP client.

Notes and tips:

• Environment variables: depending on deployment, you may need to set UPWORKDIR or UNITY_PROJECT_PATH to point the server at your Unity project location.
• Networking: ensure your MCP client can reach the server’s host/port. If using Docker, you may need to expose ports or use a host network depending on your setup.
• Data handling: the server fetches current ParticleSystem state on inquiry; for modifications, ensure changes are validated against Unity’s API expectations to avoid runtime errors.
• Compatibility: the server supports all 24 ParticleSystem modules. If you enable advanced features (e.g., Sub Emitters, External Forces, Custom Data), ensure your Unity version supports those modules.
• Troubleshooting: common issues include misconfigured project paths, insufficient permissions to read Unity assets, or mismatched Unity versions between client expectations and the project.
• Documentation: refer to the Unity particle system documentation for module-specific parameters when crafting modify payloads.