广告合作
MCP.so
登录

UML MCP Server

@Swayingleaves

代码仓库

基本信息

分类

其他

传输方式

stdio

作者

Swayingleaves

关于 UML MCP Server

暂无概览

UML MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "uml-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@Swayingleaves/uml-mcp-server",
        "--client",
        "claude"
      ]
    }
  }
}

UML MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。

概览

What is UML-MCP-Server?

UML-MCP-Server is a UML diagram generation tool based on the Model Context Protocol (MCP). It helps users create various types of UML diagrams by describing them in natural language or by directly writing PlantUML code. It is intended for developers and designers who need to generate UML diagrams quickly and integrate them into MCP-compatible clients like Claude or Cursor.

How to use UML-MCP-Server?

Installation can be done via Smithery (npx -y @smithery/cli install @Swayingleaves/uml-mcp-server --client claude) or manually (clone the repository, create a virtual environment, and install dependencies from requirements.txt). To use it as an MCP server, configure the server in your MCP client (e.g., Cursor) by editing the client’s config.json with the command uv and arguments including the absolute path to the project directory and the output directory. You can also use it as a Python library by calling generate_uml() with PlantUML code.

Key features of UML-MCP-Server

  • Supports nine UML diagram types: class, sequence, activity, use case, state, component, deployment, and object.
  • Generate diagrams from natural language descriptions.
  • Accept direct PlantUML code input.
  • Returns PlantUML code, an accessible URL link, and a local file path.
  • Allows a custom output directory for saved UML images.
  • Integrates as an MCP server with clients such as Claude and Cursor.
  • Includes a comprehensive logging system for recording server and operation logs.

Use cases of UML-MCP-Server

  • Quickly generate a class diagram by describing the system architecture in natural language.
  • Embed UML diagram generation into a code editor like Cursor for on-the-fly design documentation.
  • Produce PlantUML code and a shareable URL for collaboration without requiring local rendering tools.
  • Automatically save generated diagrams to a project-specific directory for version-controlled documentation.

常见问题

1

What UML diagram types can UML-MCP-Server generate?

It can generate class diagrams, sequence diagrams, activity diagrams, use case diagrams, state diagrams, component diagrams, deployment diagrams, and object diagrams.

2

How do I install UML-MCP-Server?

You can install it via Smithery with a single npx command, or manually by cloning the repository, setting up a Python virtual environment, and installing the dependencies from `requirements.txt`.

3

What does UML-MCP-Server return after generating a diagram?

It returns three items: the PlantUML code, a publicly accessible URL for viewing the diagram, and a local file path where the generated image is saved.

4

How do I configure UML-MCP-Server in Cursor?

Edit the Cursor configuration file (e.g., `~/Library/Application Support/Cursor/config.json` on macOS) to add an `mcpServers` entry with the command `uv` and arguments that point to the project directory and set the `output_dir`.

5

What common issues might I encounter?

Issues can include network problems reaching the PlantUML server, missing write permissions on the output directory, or port conflicts when starting the MCP server. Check the logs in the `logs` directory for error details.

评论

其他 分类下的更多 MCP 服务器

FastMCP v2 🚀

jlowin

🚀 The fast, Pythonic way to build MCP servers and clients.

26.0K

EverArt

modelcontextprotocol

Model Context Protocol Servers

88.1K

YOUB

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K

LoadOptimizer

LoadOptimizer plans truck, container and pallet loads with a real optimization engine, straight from your AI assistant. Describe the order in plain language, for example "plan 36 Euro pallets on a standard 13.6 m trailer

广告合作