Add AI to your Flutter app with marionette_mcp

MCP server enabling AI agents to interact with Flutter apps at runtime - let them inspect widgets, simulate taps, enter text, scroll, and take screenshots.

CategoryAI/ML
Stars★ 355
Last update2026-07-16
PriorityLow

marionette_mcp is an open-source AI/ML toolkit for Flutter mobile app development, with 355★ on GitHub and last updated on 2026-07-16. This guide covers what marionette_mcp does, why it matters in 2026, how to add it to your project, when to reach for it, and how it compares to the alternatives — plus a quick FAQ.

What is marionette_mcp?

MCP server enabling AI agents to interact with Flutter apps at runtime - let them inspect widgets, simulate taps, enter text, scroll, and take screenshots. It focuses on bringing on-device or cloud AI features into a Flutter app. The project lives at leancodepl/marionette_mcp and is maintained by leancodepl.

Why marionette_mcp is worth knowing in 2026

marionette_mcp carries 355 GitHub stars, 39 forks, 8 open issues. It has been around since 2026, and is actively maintained (updated within the last month). For a AI/ML option, that combination of adoption and upkeep usually means a healthy community, production usage, and plenty of examples to learn from — the things that make a dependency safe to build on.

Installing marionette_mcp

Add the package to your pubspec.yaml:

dependencies:
  marionette_mcp: ^latest

Then fetch it and import it in your Dart code:

flutter pub get
import 'package:marionette_mcp/marionette_mcp.dart';

Check the package’s example/ directory and its pub.dev page for the exact API — marionette_mcp is versioned there with full docs so you always integrate against the current release.

When should you use marionette_mcp?

Reach for marionette_mcp when:

  • you’re adding a chatbot, assistant, or generative feature
  • you need on-device inference for privacy or offline use
  • you’re wiring an LLM or ML model into a mobile app

marionette_mcp vs. the alternatives

If you’re weighing options in the AI/ML space, these are the other projects developers most often compare marionette_mcp against:

There’s no single winner — the right pick depends on your app’s size, your team’s familiarity, and your performance budget. Browse the full AI/ML collection to compare them side by side.

Frequently asked questions

Is marionette_mcp free to use?

Yes. marionette_mcp is open source and free to use in your Flutter projects. You can view the source, report issues, and contribute on GitHub.

Does marionette_mcp work on both iOS and Android?

marionette_mcp is built for Flutter, so it targets iOS and Android from a single codebase, and typically web and desktop too depending on the project’s platform support.

As of 2026, marionette_mcp has around 355 stars and 39 forks on GitHub, which puts it among the more widely used options in the AI/ML space.

What are good alternatives to marionette_mcp?

Popular alternatives in the AI/ML category include appflowy, appwrite, some-many-books. The best choice depends on your app’s size, team, and performance needs.

How do I install marionette_mcp?

Add marionette_mcp to the dependencies section of your pubspec.yaml and run flutter pub get. Full versions and API docs are on pub.dev.


Part of FlutterCook — 500 hands-on guides to the best open-source Flutter libraries, UI kits, and apps. Explore the live GitHub trends or browse YouTube guides.