PLECS-MCP

MCP serverCommunity

Wraps all 22 PLECS Standalone JSON-RPC methods as MCP tools, with model checks, plots and FFT/THD post-processing. By Supertaskchat.

2 starsUpdated 7 days agoGitHub data from 5 Nov 2026

At a glance

Type: MCP server
Works with: PLECS
How it connects: Drives a running copy of the tool
Access: Can modify models
Tool versions: PLECS Standalone with RPC enabled (22/22 methods on PLECS 5.0.7); Python 3.10 or later
Operating systems: Windows, macOS, Linux
Needs a PLECS licence: Yes
Transport: Local (stdio), Remote (HTTP)
AI clients documented: Claude Desktop, Cursor, Codex
Licence: MIT
Language: Python

What it can do

One tool per RPC method: load, get and set parameters, simulate, analyse (AC sweep, loop gain), scope control, code generation and FMU export.
Checks a model's connection graph and layout before simulating, and returns scope captures and waveform or Bode plots as inline PNG images.
Runs background simulation tasks, pass/fail checks on signals, component-library search and file-level model edits (add components, wire, set parameters).
Starts PLECS automatically on macOS, Windows or Linux when the RPC port is closed.

Limitations

PLECS only accepts RPC from the same machine, so the server must run next to PLECS.
Writes are limited to allowed folders; irreversible calls such as protect need confirm=true.
Install instructions use the author's own local paths; there is no PyPI package.

Install

cd "/Users/guohuawei/Workspace/AI Usecase/PLECS-MCP"
uv venv --python 3.13 .venv
uv pip install -e . --python .venv/bin/python

Source: github.com/Supertaskchat/PLECS-MCP#install

{
  "mcpServers": {
    "plecs": {
      "command": "/Users/guohuawei/Workspace/AI Usecase/PLECS-MCP/.venv/bin/python",
      "args": ["-m", "plecs_mcp"]
    }
  }
}

Source: github.com/Supertaskchat/PLECS-MCP#run

MCP servers and skills run code on your computer. Review the source before installing.

AI-generated models, drawings and results must be checked by a qualified engineer before use.