AutoCAD MCP

AutoCAD MCP

Enables AI clients like Claude Desktop to control AutoCAD through natural language commands using the Model Context Protocol.

Category
Visit Server

README

AutoCAD MCP

通过 MCP(Model Context Protocol)让 AI 客户端(Claude Desktop / Cursor 等)用自然语言控制 AutoCAD 绘图。参考并改进自 daobataotie/CAD-MCP(差异见 docs/references/cad-mcp-analysis.md)。

环境要求

  • Windows + 已安装 AutoCAD(本项目在 AutoCAD 2026 上开发,ProgID AutoCAD.Application.25.1,见 src/autocad_mcp/config.json
  • conda

快速开始

conda activate autocad-mcp
# 打开 AutoCAD 2026,然后:
mcp dev src/autocad_mcp/server.py    # 用 MCP Inspector 调试

接入 Claude Desktop:在 claude_desktop_config.json(Windows 上一般是 %APPDATA%\Claude\claude_desktop_config.json)里加入 mcpServers

{
  "mcpServers": {
    "autocad-mcp": {
      "command": "C:\\Users\\<用户名>\\.conda\\envs\\autocad-mcp\\python.exe",
      "args": ["-m", "autocad_mcp.server"]
    }
  }
}

command 直接指向 conda 环境里的 python.exe,不要用 conda run -n autocad-mcp ...——后者在 Claude Desktop 拉起子进程时可能拿不到 conda 的 shell 初始化环境,也更容易碰到 Windows 控制台编码问题。前提是已经 pip install -e . 把本项目装进这个环境。)

网页 UI(自己选大模型,仍走 MCP 协议)

除了接 Claude Desktop,也可以用自带的网页界面,自己填 API Key、自己选大模型(Claude / OpenAI / OpenAI 兼容的国产模型如通义千问/DeepSeek/智谱)。网页后端是一个自建的 MCP client,通过 HTTP 连接 AutoCAD MCP server——cad//tools/ 两层完全复用,不是绕开 MCP 直接调 Python 函数。

先装网页相关依赖(装进同一个 autocad-mcp conda 环境):

"C:\Users\<用户名>\.conda\envs\autocad-mcp\python.exe" -m pip install anthropic openai fastapi "uvicorn[standard]"

需要同时起两个进程(项目根目录下执行,web 包不是通过 pip 装的,靠 -m 的当前目录解析):

# 终端 1:AutoCAD MCP server,HTTP 模式,供网页后端连接(不影响 Claude Desktop 的 stdio 配置)
"C:\Users\<用户名>\.conda\envs\autocad-mcp\python.exe" -m autocad_mcp.server --http

# 终端 2:网页后端
"C:\Users\<用户名>\.conda\envs\autocad-mcp\python.exe" -m uvicorn web.backend.app:app --port 8000

打开 http://127.0.0.1:8000,选模型、填 API Key(只在当次请求里用,不落盘不写日志)、用自然语言聊天即可。这是本地单用户场景的简化实现:会话历史存在后端进程内存里(重启即丢),没有鉴权,只适合本机 127.0.0.1 访问,不要暴露到公网。

项目结构

docs/design/architecture.md

开发文档 / 日志

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured