Server profile

Unreal Engine Generative AI Support Plugin MCP Server

This Unreal Engine plugin combines generative AI API integrations with MCP-based editor control. Assistants can manipulate scene objects, work with some Blueprint features, and run editor scripts, although MCP development is inactive.

Server facts

Unreal Engine Generative AI Support Plugin MCP Server

Tech

Runtime / language
c++C++
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
675
Last commit
Data as of

Tags

  • Blueprints
  • Game Development
  • Generative Ai
  • Scene Control
  • Unreal Engine

Overview

Unreal Engine Generative AI Support Plugin brings model integrations and assistant-driven editor actions into an Unreal project. The free repository includes chat integrations for OpenAI, Anthropic, XAI, and DeepSeek, with structured output support for OpenAI. Its MCP bridge lets Claude Desktop, Claude Code, or Cursor interact with the editor. This is aimed at game developers who want both API-based AI features and an assistant able to change their development scene.

The documented MCP capabilities include spawning objects, changing their position and scale, adjusting materials, creating Blueprint types, and adding functions or variables. Python execution and console commands are also available. The feature-status list is important: Blueprint node wiring is described as buggy, while advanced scene work, widget generation, and other capabilities remain unfinished. The README explicitly says the free plugin's MCP support is not under active development. Paid plugins are presented separately, so their broader model lists should not be treated as free-repository functionality.

Setup uses a Python bridge with FastMCP and requires Unreal's Python Editor Script Plugin. The plugin can be added as a Git submodule, then enabled in the editor. MCP control itself does not require an Anthropic API key for testing through the Claude app; direct model API use has separate key requirements. Because the bridge can directly modify the project, the README recommends a controlled environment, backups, and version control before enabling it.

Install & configure

Requirements

  • Unreal Engine project with the GenerativeAISupport plugin enabled
  • Python with FastMCP installed
  • Unreal Python Editor Script Plugin enabled
  • A supported MCP client
Documented launch / install command
python /Plugins/GenerativeAISupport/Content/Python/mcp_server.py

Replace the example bridge path with its location in your project. The README documents pip install fastmcp and adding the plugin as a Git submodule. Exact MCP tool identifiers are not stated in the supplied README.

Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows: %APPDATA%\Claude\claude_desktop_config.json

Use Settings → Developer → Edit Config when available. Do not assume a Linux configuration path; check the client documentation for your platform. Remote servers use the app’s connector flow when supported, not the local stdio file shown here. Merge this example with existing settings; do not overwrite unrelated servers.

Claude Desktop configuration
{
  "mcpServers": {
    "UnrealGenAISupport": {
      "command": "python",
      "args": [
        "/Plugins/GenerativeAISupport/Content/Python/mcp_server.py"
      ],
      "env": {
        "UNREAL_HOST": "<Unreal editor host, shown as localhost>",
        "UNREAL_PORT": "<Unreal editor connection port, shown as 9877>"
      }
    }
  }
}

Read the Claude Desktop setup guide

Config file · Project: .cursor/mcp.json

User: ~/.cursor/mcp.json

The project path is relative to the workspace root. The tilde denotes your user home directory. Use Cursor’s MCP settings to confirm which configuration and scope your installed version is reading. Merge this example with existing settings; do not overwrite unrelated servers.

Cursor configuration
{
  "mcpServers": {
    "UnrealGenAISupport": {
      "command": "python",
      "args": [
        "/Plugins/GenerativeAISupport/Content/Python/mcp_server.py"
      ],
      "env": {
        "UNREAL_HOST": "<Unreal editor host, shown as localhost>",
        "UNREAL_PORT": "<Unreal editor connection port, shown as 9877>"
      }
    }
  }
}

Read the Cursor setup guide

Config file · Workspace: .vscode/mcp.json

For user scope, use the “MCP: Open User Configuration” command if available. Do not guess a user configuration path across operating systems, profiles, containers, or remote workspaces; check the client documentation. Merge this example with existing settings; do not overwrite unrelated servers.

VS Code configuration
{
  "servers": {
    "UnrealGenAISupport": {
      "type": "stdio",
      "command": "python",
      "args": [
        "/Plugins/GenerativeAISupport/Content/Python/mcp_server.py"
      ],
      "env": {
        "UNREAL_HOST": "<Unreal editor host, shown as localhost>",
        "UNREAL_PORT": "<Unreal editor connection port, shown as 9877>"
      }
    }
  }
}

Read the VS Code setup guide

Config file · macOS/Linux: ~/.codeium/windsurf/mcp_config.json

Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json

Open the MCP configuration through Windsurf’s settings when possible to confirm the active file. Paths can differ in managed or future releases; check the client documentation if this location is not present. Merge this example with existing settings; do not overwrite unrelated servers.

Windsurf configuration
{
  "mcpServers": {
    "UnrealGenAISupport": {
      "command": "python",
      "args": [
        "/Plugins/GenerativeAISupport/Content/Python/mcp_server.py"
      ],
      "env": {
        "UNREAL_HOST": "<Unreal editor host, shown as localhost>",
        "UNREAL_PORT": "<Unreal editor connection port, shown as 9877>"
      }
    }
  }
}

Read the Windsurf setup guide

Config file · Project scope: .mcp.json at the repository root

The project file uses mcpServers. For user scope, use claude mcp add --scope user rather than assuming .mcp.json in your home directory is the user configuration. Check the client documentation for managed settings and storage details. Merge this example with existing settings; do not overwrite unrelated servers.

Claude Code configuration
{
  "mcpServers": {
    "UnrealGenAISupport": {
      "type": "stdio",
      "command": "python",
      "args": [
        "/Plugins/GenerativeAISupport/Content/Python/mcp_server.py"
      ],
      "env": {
        "UNREAL_HOST": "<Unreal editor host, shown as localhost>",
        "UNREAL_PORT": "<Unreal editor connection port, shown as 9877>"
      }
    }
  }
}

Read the Claude Code setup guide

Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.

Tools

No individual tool names were confirmed in the source material. Check the repository for its current tool list.

Names, permissions, and defaults can change. Check the current upstream documentation before granting access.

Use cases

  • Let an assistant spawn and transform objects in an Unreal scene.
  • Create Blueprint types and add functions or variables.
  • Run Python scripts or console commands through an assistant.
  • Integrate supported chat APIs into Unreal gameplay experiments.

Strengths

  • Combines Unreal editor control with model API integration.
  • Supports Claude Desktop, Claude Code, and Cursor.
  • The free plugin is MIT-licensed.

Limitations

  • MCP support in the free plugin is not actively developed.
  • Several advertised feature areas are unfinished or buggy.
  • Direct project control requires careful backups and isolation.

Frequently asked questions

The README says no key is needed to test MCP through the Claude app. Calling the Anthropic API is a separate use case that needs a key.

No. Type creation and function or variable additions are marked complete, but node connections are described as buggy and advanced generation is unfinished.

No. The README separates the free repository's supported integrations from its paid plugin offerings.

Use MCP in a controlled environment and back up the project with version control because the assistant can directly change it.

Independent listing, not an affiliation or security review. Grant the least access needed and verify the upstream instructions before running code.

Related MCP servers

BlenderMCP

Games & 3D

BlenderMCP links AI assistants to Blender for scene inspection, object and material changes, and Python execution. Optional asset integrations add Poly Haven downloads, Sketchfab models, and generated 3D assets.

MCP for Unity

Games & 3D

MCP for Unity links AI assistants to the Unity Editor for scene work, asset management, script editing, testing, and builds. It targets game developers who want to direct editor workflows through natural language.

Godot MCP

Games & 3D

Godot MCP lets AI agents open the editor, run projects, and inspect debug output. It also supports scene and node operations, sprite loading, mesh-library export, and UID maintenance for supported Godot versions.

Unreal MCP

Games & 3D

Unreal MCP links assistant requests to Unreal Editor through a Python server and C++ plugin. It covers scene objects, Blueprint construction, and viewport adjustments, but the project explicitly remains experimental.

MCP Unity Editor

Games & 3D

Connect AI assistants to Unity Editor through a Unity package and Node.js MCP bridge. Manage scenes, GameObjects, materials, packages, tests, and console logs, with resources and guided workflow prompts.

Minecraft MCP Server

Games & 3D

Minecraft MCP Server lets an AI assistant control a Minecraft character through Mineflayer. Tools cover movement, blocks, inventory, smelting, entities, and chat, with a documented local LAN-world setup.