UEMCP

UEMCP is a two-tier MCP setup that brokers AI-assisted Unreal Engine workflows. The Node.js MCP server exposes a collection of specialized tools that wrap common Unreal Engine Python API calls, enabling AI agents to perform tasks such as spawning actors, capturing viewport images, querying project data, and organizing level content without writing bespoke Python code. The accompanying Python Editor Plugin provides full Python access within the UE editor via the python_proxy capability, allowing the AI to execute arbitrary Python when needed while the MCP tools offer optimized, safer shortcuts for frequent operations. Use the CLI-visible tools to perform standard tasks (e.g., actor_spawn, viewport_screenshot, asset_list) and rely on python_proxy for advanced automation or custom UE Python workflows.

Prerequisites:

• Node.js (LTS recommended) and npm
• Python 3.11+ (for the editor plugin and tooling)
• Access to a Unreal Engine project (optional for testing in-editor workflows)

Install and run:

1. Clone the repository: git clone https://github.com/atomantic/UEMCP.git cd UEMCP

2. Install Node dependencies for the MCP server: npm install

3. Build or prepare the server (if a build step exists in this repo): npm run build

4. Start the MCP server: npm start

5. (Optional) If you need the editor plugin and setup scripts, follow the README’s guidance to run setup.sh from the repo root and install the plugin into your UE project.

Note: The setup script can configure AI development tools (Claude, Amazon Q, Gemini, OpenAI Codex) and detect your Unreal project path. Ensure Unreal Engine is installed and your environment paths are accessible to the startup script.

Notes and tips:

• The MCP server operates as a Node.js process; ensure your environment allows long-running processes and network access for remote AI clients.
• Python integration via python_proxy enables full UE Python API access within the editor; use dedicated tools for common tasks to reduce complexity and improve reliability.
• If you customize or fork UEMCP, the architecture is designed for extension; add new MCP tools by wrapping UE Python calls in concise tool functions.
• Common issues include path mismatches to the Unreal project, missing Python dependencies in the editor plugin, or environment variables required by specific AI tool integrations (Claude, Gemini, etc.).
• Check logs for errors related to UE plugin installation, tool availability, or editor integration.