Skip to main content
Glama

UE Agent Harness

UE5.8 编辑器的 AI Agent 中间层。Harness 作为 MCP Server 面向 LLM(Claude/GPT),作为 MCP Client 面向 UE MCP Server,提供上下文组装、状态缓存、视觉验证闭环、长任务记忆压缩、安全护栏。

快速开始

# 1. 在 UE 编辑器中启用 MCP Server
#    Editor Preference → General → Model Context Protocol → Auto Start Server

# 2. 启动 Harness
harness start --ue-port 8000 --listen-port 9000

# 3. 在 Claude Code 或其他 MCP 客户端中连接到 localhost:9000

Related MCP server: Monolith

文档

项目结构

ue-agent-harness/
├── harness/              # Python 包
│   ├── cli.py            # CLI 入口
│   ├── server.py         # MCP Server(面向 LLM)
│   ├── client.py         # MCP Client(面向 UE)
│   ├── config.py         # 配置管理
│   ├── context/          # 上下文组装
│   ├── verification/     # 视觉验证
│   ├── state/            # Write-Through State Cache
│   ├── memory/           # 任务记忆压缩
│   ├── recovery/         # 错误恢复
│   ├── safety/           # 安全护栏
│   └── observability/    # 日志与回放
├── skills/               # Harness Skill 示例
├── tests/                # 测试
├── docs/                 # 文档
│   ├── architecture.md
│   ├── CONTEXT.md
│   ├── adr/              # 架构决策记录
│   └── issues/           # 开发 Issue
└── pyproject.toml
F
license - not found
-
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/MDX-0003/UE-MCP-Harness'

If you have feedback or need assistance with the MCP directory API, please join our Discord server