Figma to Flutter MCP Server

Figma to Flutter MCP Server

Enables AI coding agents to extract Figma design data and convert it into Flutter widgets and screens. Supports theme setup, component analysis, asset exports, and provides Flutter implementation guidance from Figma designs.

Category
Visit Server

README

<div align="center"> <img src="docs/images/figma-flutter-mcp.png" alt="Theme Setup Example" style="max-width: 100%; height: auto;">

<br>

<h1>Figma to Flutter MCP Server</h1> <p> 🌐 Available in: <a href="README.ko.md">한국어 (Korean)</a> | <a href="README.ja.md">日本語 (Japanese)</a> | <a href="README.zh-cn.md">简体中文 (Simplified Chinese)</a> | <a href="README.zh-tw.md">繁體中文 (Traditional Chinese)</a> </p> <h3>Utilize Figma's rich data in your coding agent.<br/>Implement designs in Flutter way!</h3> <a href="https://npmcharts.com/compare/figma-flutter-mcp?interval=30"> <img alt="weekly downloads" src="https://img.shields.io/npm/dm/figma-flutter-mcp.svg"> </a> <a href="https://github.com/mhmzdev/figma-flutter-mcp/blob/main/LICENSE"> <img alt="MIT License" src="https://img.shields.io/github/license/mhmzdev/figma-flutter-mcp" /> </a> <a href="https://twitter.com/mhmzdev"> <img alt="Twitter" src="https://img.shields.io/twitter/url?url=https%3A%2F%2Fx.com%2Fmhmzdev&label=%40mhmzdev" /> </a> </div> <br>

Use Cursor or other AI-powered tools to access Figma's rich files, data, components and much more using MCP server.

📋 Table of Contents

🎥 Video Demo

Showcased almost all the features of Figma Flutter MCP with real figma design.

  • English: https://youtu.be/o_GezwAlXlU
  • Urdu/Hindi: https://youtu.be/mepPWpIZ61M

📝 Getting Started

You may explore the detailed getting started docs or the demo video as quick-start. As its a First Release hence there's a lot of room for improvements so you can checkout the issues to see what else there's to work or to improve.

📚 How it works | Details Here

  1. Components/Widgets
  • ✅ Extract Figma node data: Layout, styling, dimensions, colors, text content, etc.
  • ✅ Analyze structure: Child elements, nested components, visual importance
  • ✅ Provide guidance: Suggest Flutter widgets and implementation patterns
  • ❌ NOT generating actual Flutter code files
  1. Screens
  • ✅ Extract screen metadata: Device type, orientation, dimensions
  • ✅ Identify sections: Header, footer, navigation, content areas
  • ✅ Analyze navigation: Tab bars, app bars, drawers, navigation elements
  • ✅ Provide Scaffold guidance: Suggest Flutter screen structure
  • ❌ NOT generating actual Flutter screen

Since its just helping AI write Flutter code so it means the better your prompt will be the better results you'll get.

🛠️ Usage

Following steps shows a minimal usage and setup instructions:

🔑 Figma API Key

You will need to create a Figma access token to use this server. Instructions on how to create a Figma API access token can be found here.

🏹 MCP in Cursor

Once you've the FIGMA API KEY, you can setup the MCP in cursor as follows:

  1. Press CMD + Shift + P (Ctrl on Windows)
  2. Type "Open MCP Settings"
  3. Click on "Add new MCP"
  4. Paste the below json object

MacOS/Linux

{
  "mcpServers": {
    "Figma Flutter MCP": {
      "command": "npx",
      "args": ["-y", "figma-flutter-mcp", "--figma-api-key=YOUR-API-KEY", "--stdio"]
    }
  }
}

Windows

{
  "mcpServers": {
    "Figma Flutter MCP": {
      "command": "cmd",
      "args": ["/c", "npx", "-y", "figma-flutter-mcp", "--figma-api-key=YOUR-API-KEY", "--stdio"]
    }
  }
}

NOTE: If you've installed this MCP as npm package make sure to keep it updated to latest version. Sometimes, it caches the old version and keep showing you error like "Not being able to use tool call" or "Figma API key setup is not working" etc.

🚀 Quick Start for Local Testing

Prerequisites

  • Node.js 18+
  • Figma API Key (Access Token)
  • Cursor AI IDE with MCP support
  • Flutter SDK

For quick local testing, you can run the server via HTTP instead of stdio:

# Clone and setup
git clone <your-repo-url> figma-flutter-mcp
cd figma-flutter-mcp
npm install

# Create .env file with your Figma API key
echo "FIGMA_API_KEY=your-figma-api-key-here" > .env

# Start HTTP server for local testing
npm run dev

Then add this to your MCP client configuration:

{
  "mcpServers": {
    "local-figma-flutter": {
      "url": "http://localhost:3333/mcp"
    }
  }
}

See CONTRIBUTING.md for detailed instructions.

🧱 Basic Workflow

🤖 AI Coding Agent Assistance

For better results you can setup some instructions in following files as per your AI Coding Agent:

  • Cursor: .cursor/rules/fluttering.mdc
  • Claude: CLAUDE.md
  • Gemini CLI: GEMINI.md

This way your AI agent will use the MCP's output and ensure the flutter code is as per your project requirements and structure. You can checkout an example of cursor rules that I used for testing this out.

  1. Setup Theme & Typography: The most efficient way, put two frames in Figma with Theme colors and Typography samples on it. For instance:

Theme Setup Example Typography Setup Example

  • Figma Desktop: Select the frame and press CMD + L or Ctrl + L
  • Figma Web: Select the frame and copy the URL

💡 HINT: The valid URL will contain a FILE ID and NODE ID params

"Setup flutter theme from <figma_link> including Colors and Typography.
  1. Widget Generation: The most efficient way, use COMPONENTS in figma. For example:

Button

This one has 8 variants, you may prompt whether you want to have variants or not.

"Create this widget in flutter <figma_link>, setup only 2 variants for now and break the files in smaller parts for code readability.

If you do not have COMPONENTS in figma, you can use FRAME just prompt the AI that you want this to be a widget and it will handle the rest.

  1. Full Screen Generation: If there are any IMAGE ASSETS available, it will export them and put them in assets/ along with pubspec.yaml
"Create full screen from this figma link <figma_link>, ensure the code is readable by having smaller files
  1. Assets Export:
  • Image Assets: Will work automatically when generating screens
"Export this image asset from figma <figma_link>
  • SVG Assets: Will NOT work automatically, explained below.
"Export this as an SVG asset from Figma: <figma_link>

⚠️ Why SVG assets don’t work with screen generation

clarify that vectors include icons and pen-tool shapes, so bulk exports may grab unintended nodes; recommend exporting SVGs separately. This process still saves you a lot of time by exporting them in your assets/svg/ directory and updating your pubspec.yaml.

🧰 MCP Tools

Assets related:

  • export_flutter_assets: Individual tool for image assets used with Screen generation
  • export_svg_flutter_assets: Individual tool for SVG assets export

Widget related:

  • analyze_figma_component: For type=COMPONENT in figma or FRAME prompted by user
  • list_component_variants: For type=COMPONENT_SET in figma (Widget variants)
  • inspect_component_structure: For nested COMPONENTS or FRAMES

Full screen related:

  • analyze_full_screen: type=FRAME for full screen and assets export (image only)
  • inspect_screen_structure: For layouts and other information for building screen

⚠️ Disclaimers

  • Figma Design: Since we're using Figma's API to fetch the node and its details, so the better design you have the more better it will interpret for the AI to consume i.e. auto layouts, frame usage over group usage, consistently aligned across the board.
  • Use Case: At this stage, its highly recommend to NOT use it to develop scalable apps rather try and play it with MVPs, smaller and explanatory tasks.
  • Rate limiting: Heavy usage may trigger Figma rate limits (e.g., HTTP 429). The server includes retry with backoff, but it does not bypass Figma limits. If you encounter rate limits, wait a few minutes and reduce the request volume.

🙌🏼 Acknowledgments

I came across Figma Context MCP by Graham Lipsman that sparks this motivation for me to develop Figma to Flutter explicitly having features like:

  • Assets exports
  • Colors and Theme setups
  • Widget tree and full screen building

Others coming soon...

🧱 Other framworks

If you want to develop this for React, Angular, React Native, Vue or any other framework. I've added a detailed doc Figma Framework MCP that you can explore and get started. Meanwhile I'll maintain a list here if someone's already doing this for framework specific Figma's MCP servers.

  • ...
  • ...

🔑 License

This project is licensed under the MIT License - see the LICENSE file for details

🙋‍♂️ Author

Muhammad Hamza

LinkedIn Link

You can also follow my GitHub Profile to stay updated about my latest projects:

GitHub Follow

If you liked the repo then kindly support it by giving it a star ⭐!

Copyright (c) 2025 MUHAMMAD HAMZA


Built with ❤️ for designers and developers who want to bridge the gap between design and code.

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