UE5
MCP for Unreal Engine 5
claude mcp add --transport stdio vedantrgosavi-ue5-mcp python -m ue5_mcp \ --env UE_PATH="Path to Unreal Engine 5 installation (e.g., /Applications/UE_5.x/Engine/Binaries/UE4Editor.app/Contents/MacOS/UE4Editor or equivalent on Windows/Linux)" \ --env PYTHON_VENV="Optional: path to a Python virtual environment to isolate dependencies" \ --env BLENDER_PATH="Path to Blender executable (e.g., /usr/local/bin/blender or C:\Program Files\Blender Foundation\Blender.exe)" \ --env REQUIREMENTS_FILE="Optional: path to requirements.txt if not using default in repo (e.g., ./requirements.txt)"
How to use
UE5-MCP provides an end-to-end AI-assisted workflow bridge between Blender and Unreal Engine 5. It builds upon BlenderMCP to enable text- and image-driven scene generation, asset management, and automated UE5 integration. The server orchestrates Python-based tooling that coordinates Blender-side scene creation, material and asset management, and Unreal Engine-side import, level design automation, and blueprint-assisted debugging. Use UE5-MCP to kick off AI-driven workflows that generate scenes, transfer assets, and streamline level creation between Blender and UE5. The documentation notes that you can run MCP within Blender or UE5, leveraging the provided commands and configuration to trigger AI-powered automation features, aided by dependencies listed in requirements.txt.
How to install
Prerequisites:
- Python 3.x installed
- Blender installed (latest stable)
- Unreal Engine 5 installed
- Access to repository with UE5-MCP code
Installation steps:
- Clone the repository: git clone https://github.com/your-repo/UE5-MCP.git
- Create and activate a Python virtual environment (optional but recommended): python -m venv venv source venv/bin/activate # on macOS/Linux venv\Scripts\activate # on Windows
- Install dependencies from requirements.txt: pip install -r requirements.txt
- Configure settings (see configurations.md in the repo) and ensure Blender and UE5 paths are correctly set in environment variables if necessary.
- Run the MCP server (see the mcp_config section for invocation details). You can run it from the repository root using the configured Python module: python -m ue5_mcp
- Follow workflow-specific instructions (workflow.md) to perform Blender-to-UE5 automation tasks.
Additional notes
Tips and known considerations:
- Ensure Blender and UE5 paths are correctly set in environment variables when running the MCP server to enable cross-application automation.
- If you encounter dependency issues, use a virtual environment and confirm the requirements.txt matches your Python version.
- The project mentions multiple workflow and documentation files (configurations.md, workflows.md, commands.md, troubleshooting.md); consult these for detailed commands and parameter usage.
- Because the project was not fully finalized, some features may be experimental. Validate AI-generated content in UE5 and Blender prior to production use.
- If Blender or UE5 tool integrations fail to initiate, check that their respective command-line interfaces are accessible from your shell and that required plugins or scripts are enabled.
Related MCP Servers
Unreal_mcp
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to control Unreal Engine through the native C++ Automation Bridge plugin. Built with TypeScript, C++, and Rust (WebAssembly) for ultra-high-performance game development automation.
robloxstudio
Create agentic AI workflows in ROBLOX Studio
okta
The Okta MCP Server is a groundbreaking tool built by the team at Fctr that enables AI models to interact directly with your Okta environment using the Model Context Protocol (MCP). Built specifically for IAM engineers, security teams, and Okta administrators, it implements the MCP specification to help work with Okta enitities
zerodha
Zerodha MCP Server & Client - AI Agent (w/Agno & w/Google ADK)
unity-api
Instant, accurate Unity API lookups instead of expensive source file reads, saving your agent tokens, context, and hallucinations
midnight
Midnight MCP server giving AI assistants access to Midnight blockchain — search contracts, analyze code, explore docs