AutoCAD Electrical MCP
MCP server and web dashboard for controlling AutoCAD Electrical 2027 via natural language, supporting MCC configurator and multiple AI backends like Claude, Ollama, or OpenAI-compatible models.
README
AutoCAD Electrical Interface (ECM)
Local web dashboard for AutoCAD Electrical 2027 - control AutoCAD with plain language using Claude, Ollama or any OpenAI compatible models. Comes with a motor control center (MCC) configurator panel to edit or create complete MCC drawing packages.
This app is designed for Electrical Controls Manufacturing (ECM), aimed at adding support for their Motor Control Center (MCC) products, as well as planned future support for variable frequency drives (VFDs), soft starters, breaker panels, and other products. The MCC configurator will not work unless MCC_LAYOUT, MCC_UNITDATA, and MCC_NAMEPLATE.dwg are open in AutoCAD, and the required CAD blocks are located in the filepath written in .env. Additionally, the general data configurator panel will not work unless the 'General_Data Sheet.dwg' file is open and contains the 'GENRA001B' block. The original repo used for this project is available at https://github.com/Igualguana/AUTOCAD-ELECTRICAL-MCP.
MCC Configurator Setup
The MCC Builder requires three template drawings to be open in AutoCAD before you start a project. These are the files the tool writes into — they must come from your company's block library.
| File | Purpose |
|---|---|
MCC_LAYOUT.dwg |
Receives section frames and unit blocks |
MCC_UNITDATA.dwg |
Receives UDATALIN data rows |
MCC_NAMEPLATE.dwg |
Receives LAMACOID nameplate rows |
General_Data Sheet.dwg |
(Optional) Receives GENRA001B data rows |
Steps:
- Set
MCC_BLOCK_LIBRARYin your.envto the folder containing these files (and all unit block DWGs). - Open
MCC_LAYOUT.dwg,MCC_UNITDATA.dwg, andMCC_NAMEPLATE.dwgin AutoCAD Electrical. - (Optional) Open
General_Data Sheet.dwgin AutoCAD Electrical. - Start the web server:
python start_web.py - Open
http://127.0.0.1:8080and click MCC in the sidebar. - Click New Project — the tool will connect to the three open drawings automatically.
If you see "MCC_LAYOUT.dwg is not open" when creating a project, go back to step 2 and make sure the required files are open in AutoCAD before clicking New Project.
Web Interface
The web dashboard is a FastAPI + vanilla JS single-page app served locally at http://127.0.0.1:8080.
Running locally
# Start the web server (opens browser automatically)
python start_web.py
# Or specify host/port explicitly
python start_web.py --host 127.0.0.1 --port 8080 --no-browser
Requirements
- Windows 10/11 (64-bit) — COM automation is Windows-only
- Python 3.11+
- AutoCAD Electrical 2027 running with the required .dwg files open
- Mode A only: Anthropic API key
- Mode B only: Ollama installed and running
Installation
Option A — One-click installer (recommended)
- Download or clone the repo — if you don't have Git yet, download the ZIP from GitHub and unzip it.
- Open the folder, right-click
install.bat→ Run as administrator.
The script will:
- Install Python 3.11 via
wingetif it isn't already on your machine - Install Git via
wingetif it isn't already on your machine - Install all required Python libraries automatically
- Copy
.env.example→.envso you can fill in your settings
wingetis built into Windows 10 (1709+) and Windows 11. If your machine doesn't have it, install Python 3.11+ and Git manually, then run Option B below.
Option B — Manual install
# 1. Clone
git clone https://github.com/nihaalgaina/AutoCAD-Electrical-Interface-ECM
cd AutoCAD-Electrical-Interface-ECM
# 2. Install dependencies
python scripts/install.py
Configure your environment
Open .env in any text editor and fill in the values for your setup:
# Path to the folder containing your MCC block .DWG files
# (MCC_400.DWG, UNIT4040.DWG, UDATALIN.DWG, LAMACOID.DWG, etc.)
MCC_BLOCK_LIBRARY=C:\Path\To\Your\MCC Blocks
# Only needed if you want to use Claude as the AI provider
ANTHROPIC_API_KEY=sk-ant-...
# Only needed for OpenAI / Groq
OPENAI_API_KEY=...
GROQ_API_KEY=...
MCC_BLOCK_LIBRARYis the only setting most users need to change. Point it at whichever folder holds your company's.DWGblock library. Multiple folders are supported — separate them with a semicolon:MCC_BLOCK_LIBRARY=C:\Blocks\400mm;C:\Blocks\500mm
Ollama setup (Mode B — local AI, no API key required)
# Install from https://ollama.com then:
ollama serve # start the server
ollama pull qwen2.5:0.5b # fast, runs on 4 GB RAM
# or
ollama pull qwen2.5:7b # better quality, needs ~6 GB RAM
Recommended:
qwen2.5:0.5b(397 MB) works well for the keyword pre-router and is reliable on machines with limited RAM. For the full LLM path useqwen2.5:7bor larger.
License
This project is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0).
You are free to use, modify, and distribute this software. Any modified version run as a network service must make its full source code available to users of that service.
See the LICENSE file for the complete license text.
Author (Original Repository)
Randy Igualguana
Copyright © 2026 Randy Igualguana
Original Repo: https://github.com/Igualguana/AUTOCAD-ELECTRICAL-MCP
Built with Model Context Protocol · FastMCP · FastAPI · pywin32 · Ollama
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.