UE-MCP
Gives AI assistants deep read/write access to the Unreal Editor through 21 category tools covering 525+ actions, plus a YAML flow engine for multi-step workflows.
README
UE-MCP
Unreal Engine Model Context Protocol Server - gives AI assistants deep read/write access to the Unreal Editor through <!-- count:tools -->21<!-- /count --> category tools covering <!-- count:actions -->525+<!-- /count --> actions, plus a YAML flow engine for multi-step workflows.
flowchart LR
AI[AI Assistant] -->|stdio| MCP[MCP Server<br/>TypeScript / Node.js]
MCP -->|WebSocket<br/>JSON-RPC| Plugin[C++ Bridge Plugin<br/>inside Unreal Editor]
Plugin -->|UE API| Editor[Editor Subsystems]
MCP -->|direct fs| FS[Config INI<br/>C++ Headers<br/>Asset Listing]
Blueprints, materials, levels, actors, animation, VFX, landscape, PCG, foliage, audio, UI, physics, navigation, AI, GAS, networking, sequencer, build pipeline — all programmable through natural language.
Quick Start
npx ue-mcp init
The interactive setup will:
- Find your
.uproject(auto-detects in current directory) - Let you choose which tool categories to enable
- Deploy the C++ bridge plugin to your project
- Enable required UE plugins (Niagara, PCG, GAS, etc.)
- Detect and configure your MCP client (Claude Code, Claude Desktop, Cursor)
Restart the editor once after setup to load the bridge plugin. To update later: npx ue-mcp update
Then ask your AI:
project(action="get_status") — verify connection
level(action="get_outliner") — see what's in the level
asset(action="list") — browse project assets
Manual Configuration
If you prefer to configure manually, add to your MCP client config:
{
"mcpServers": {
"ue-mcp": {
"command": "npx",
"args": ["ue-mcp", "C:/path/to/MyGame.uproject"]
}
}
}
Documentation
- Getting Started — Installation, configuration, first run
- Architecture — How the pieces fit together
- Tool Reference - All <!-- count:tools -->21<!-- /count --> tools with <!-- count:actions -->525+<!-- /count --> actions
- Flows - YAML flow engine, custom tasks, rollback, hooks
- Configuration —
.ue-mcp.jsonand MCP client config - Neon Shrine Demo — Interactive guided demo
- Feedback — Agent feedback system
- Troubleshooting — Common issues and fixes
- Development — Building, testing, contributing
What Can It Do?
| Category | Examples |
|---|---|
| Levels | Place/move/delete actors, spawn lights and volumes, manage splines, actor bounds |
| Blueprints | Read/write graphs, add nodes, connect pins, compile, CDO property access |
| Materials | Create materials and instances, author expression graphs |
| Assets | CRUD, import meshes/textures/animations, datatables, mesh bounds/collision/nav |
| Animation | Anim blueprints, montages, blendspaces, skeletons |
| VFX | Niagara systems, emitters, parameters |
| Landscape | Sculpt terrain, paint layers, import heightmaps |
| PCG | Author and execute Procedural Content Generation graphs |
| Gameplay | Physics, collision, navigation, navmesh inspection, behavior trees, EQS, perception, PIE damage |
| GAS | Gameplay Ability System — attributes, abilities, effects, cues |
| Networking | Replication, dormancy, relevancy, net priority |
| UI | UMG widgets, editor utility widgets and blueprints, runtime delegate inspection |
| Editor | Console, Python, PIE, viewport, sequencer, build pipeline, logs |
| Reflection | Class/struct/enum introspection, gameplay tags |
Supported Platforms
- Windows — UE 5.4–5.7
- Linux — UE 5.6+ (contributed by @robinduckett)
Requires PythonScriptPlugin (ships with UE 4.26+).
If you clone this repo to contribute, install git-lfs first - the bundled test project stores .uasset / .umap via LFS and plain git clone will leave them as pointer files.
Contributing
Issues and pull requests welcome. If an AI agent had to fall back to execute_python during your session, it will offer to submit structured feedback automatically — this helps us prioritize which native handlers to add next.
License
UE-MCP is licensed under the Business Source License 1.1 with a commercial license available for production use outside the Additional Use Grant.
- Individuals, students, and educational institutions use UE-MCP free under BUSL-1.1's Additional Use Grant. Each release converts to Apache 2.0 after four years. See LICENSE.
- Game studios, publishers, contract developers, and commercial entities using UE-MCP in proprietary products, internal pipelines, or paid services require a commercial license. See COMMERCIAL-LICENSE.md and ue-mcp.com/pricing.
Contributions are accepted under the terms of the Contributor License Agreement. See CLA.md.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。