Skip to content
Tresca

OpenModelica MCP servers and AI agent skills

OpenModelica's one listing is modelica-skills, a Claude Code plugin with 13 Modelica modelling skills and an MCP server that checks, simulates and runs physics gates on models through the OpenModelica compiler (omc), locally or in a Docker container. No standalone OpenModelica MCP server was found, and the OpenModelica project publishes none. The plugin also has an optional Dymola backend, and without an engine it falls back to static checks.

  • No official Open source server yet.
  • 1 MCP server listed.
Open source1 serverChecked 4 Nov 2026

Integrations for OpenModelica

Community and commercial servers

PluginCommunity

modelica-skillsby faveric

A Claude Code plugin with 13 Modelica modelling skills and an MCP server that checks and simulates models in OpenModelica.

LocalLinux, macOS, WindowsCan modify models4· updated 4 days ago

Listings are sorted by origin, then by recent activity. GitHub data is refreshed nightly. Nothing here is ranked.

How AI connects to OpenModelica

An MCP server or skill can only do what OpenModelica lets programs do. OpenModelica exposes 2 automation interfaces:

  • OMPython (Python): OMPython is OpenModelica's Python session handler. It talks to the OpenModelica compiler over ZeroMQ, either starting omc as a child process or connecting to one already running, and its ModelicaSystem class compiles a model, sets parameters, simulates and linearises it into A, B, C and D matrices. A runner mode replays precompiled models without omc. Documentation
  • Scripting API (.mos and omc) (Command line): The compiler also takes scripting commands written in Modelica syntax: loadModel loads a library, checkModel reports variable and equation counts, and buildModel translates a model to C and builds a simulation executable. The modelica-skills gate runner calls omc for its build, unit-checking and energy-balance gates, finding it on PATH, in standard install folders or in a running Docker container. Documentation

What Open source offers

The OpenModelica website and user's guide cover the compiler, OMPython and the scripting API, but no AI assistant, agent or MCP server as of this check. The only AI integration found is modelica-skills, a community plugin that uses OpenModelica as its simulation engine for verification gates.

Checked 4 Nov 2026. Sources: OpenModelica, OpenModelica User's Guide

What you can do with OpenModelica and AI

  • Interview you about a model's purpose, physics, ranges and sources, then split the work into tasks and build one task at a time.
  • Run build, unit-checking and energy-balance gates on a model through omc and record what each check showed.
  • Tie every correlation in a model to a cited source and check the transcription.
  • Calibrate parameters and diagnose a model that won't simulate or gives the wrong answer.
  • Review a model change independently and keep a credibility record that says no more than the evidence shows.

Taken from the listings' own documentation. Check each listing for exactly what it supports.

Installing

The listings here document setup for Claude Code and Google Antigravity. Install steps, copied from each project's own documentation, are on each listing's page. Most OpenModelica integrations run on your own computer next to OpenModelica, so you need OpenModelica installed and licensed on the same machine unless the listing says otherwise.

Questions

Is there an OpenModelica MCP server?

Only inside a plugin. modelica-skills bundles a modelica MCP server with engine_status, check_model, simulate, run_gates and check_record tools that drive OpenModelica. No standalone server was found.

How do I install it for Claude Code?

Add the plugin marketplace with claude plugin marketplace add faveric/faveric-modelica-skills, then claude plugin install modelica-skills@faveric-modelica-skills. The server needs uv, and simulation gates need omc locally or in Docker.

Can I use the skills without OpenModelica?

Yes, in static-only mode. Gates that need an engine report skip, and earlier engine results are kept for files that haven't changed.

Which OpenModelica version does it use?

The README suggests the Docker image openmodelica/openmodelica:v1.24.4-minimal, or a local install from the OpenModelica PPA on Ubuntu, the Windows installer, MacPorts or Homebrew. The adapter finds omc on PATH or through OMC_PATH.

Does the plugin change its own skills?

Only with your approval. After an issue is concluded, a learner proposes at most three lessons, and an approval skill lets you accept, edit or reject each before it is added to your library's overlay; the plugin's own skills are never rewritten.

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

Sources (checked 4 Nov 2026): OMPython (OpenModelica User's Guide) · OpenModelica scripting API · faveric/faveric-modelica-skills README · OpenModelica product page

OpenModelica is developed by its own project (Open source). Tresca is independent and not affiliated with or endorsed by the OpenModelica project.