MCP Flutter
MCP Server and Toolkit for Flutter AI Agent Driven Development - via closed feedback loop (visual & semantic snapshot) and high client side customization adaptable for any Flutter app.
What is MCP Flutter?
MCP Flutter is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server and toolkit for flutter ai agent driven development - via closed feedback loop (visual & semantic snapshot) and high client side customization adaptable for any flutter app.
MCP Server and Toolkit for Flutter AI Agent Driven Development - via closed feedback loop (visual & semantic snapshot) and high client side customization adaptable for any Flutter app.
This server falls under the Browser Automation category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- MCP Server and Toolkit for Flutter AI Agent Driven Developme
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx mcp-flutterConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MCP Flutter
MCP Flutter is a comprehensive AI-agent development toolkit for Flutter that exposes 27 MCP tools covering app inspection, interaction, debugging, and lifecycle management. It implements a closed feedback loop — AI agents can issue commands (tap, scroll, type, navigate), immediately capture a visual or semantic snapshot of the result, and iterate without human intervention. The toolkit supports dynamic tool registration, meaning Flutter apps can define and expose their own custom MCP tools at runtime, adapting the AI's capabilities to application-specific needs.
Prerequisites
- Flutter SDK installed with a working debug build (flutter run --debug)
- Dart runtime (included with Flutter SDK)
- The flutter-mcp-toolkit binary installed via the provided curl install script
- Claude Desktop, Claude Code, Cursor, Cline, or Codex as your MCP client
Install the flutter-mcp-toolkit binary
Run the official install script to set up the flutter-mcp-toolkit CLI, which powers the MCP Flutter server.
curl -fsSL https://raw.githubusercontent.com/Arenukvern/mcp_flutter/main/install.sh | bashAdd MCP integration code to your Flutter app
From your Flutter project root, run codegen-init to scaffold the MCP registration code inside your app. This is required for the server to connect to the running app.
cd my-flutter-app
flutter-mcp-toolkit codegen-initInstall agent skills
Run the init command for your target AI agent to install the skill files that describe the Flutter MCP tools. Use 'all' to install for every supported agent.
flutter-mcp-toolkit init all
# or for a specific agent:
flutter-mcp-toolkit init cursorRun your Flutter app in debug mode
Start your Flutter app in debug mode. The MCP toolkit connects via the VM service that Flutter exposes during debug builds.
flutter run --debugVerify tool registration and start an AI session
In your MCP client, ask the AI to call fmt_list_client_tools_and_resources to confirm the connection is established and see which tools (including any custom ones) are registered.
MCP Flutter Examples
Client configuration
The flutter-mcp-toolkit init command writes the config automatically. A representative Claude Desktop entry looks like this.
{
"mcpServers": {
"mcp-flutter": {
"command": "flutter-mcp-toolkit",
"args": ["serve"]
}
}
}Prompts to try
Use these prompts with Claude or another AI agent while your Flutter app runs in debug mode.
- "Call fmt_list_client_tools_and_resources to show me all registered MCP tools for this Flutter app"
- "Take a semantic snapshot of the current screen and identify any layout overflow issues"
- "Tap the 'Add to Cart' button, wait for the animation, then take a screenshot to confirm the cart count updated"
- "Evaluate the Dart expression 'MyApp.cartProvider.items.length' and report the value"
- "Hot-restart the app and compare the widget tree before and after"Troubleshooting MCP Flutter
Server connects but fmt_* tools fail with VM service errors
Ensure the Flutter app was started with 'flutter run --debug'. Release and profile builds do not expose the VM service. Also verify codegen-init was run and the generated code was compiled into the build.
Dump RPC tools produce excessive token output
Dump RPCs are disabled by default. If you explicitly enabled them with the --dumps flag and are seeing token exhaustion, remove the flag. For large widget trees, use the semantic snapshot tools instead of raw dump tools.
Custom app tools registered via dynamic registration are not visible to the AI
Call fmt_list_client_tools_and_resources first to force a refresh of the tool list. Dynamic tools are registered at runtime and may not appear until after the app has finished its initialization sequence.
Frequently Asked Questions about MCP Flutter
What is MCP Flutter?
MCP Flutter is a Model Context Protocol (MCP) server that mcp server and toolkit for flutter ai agent driven development - via closed feedback loop (visual & semantic snapshot) and high client side customization adaptable for any flutter app. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MCP Flutter?
Follow the installation instructions on the MCP Flutter GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with MCP Flutter?
MCP Flutter works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MCP Flutter free to use?
Yes, MCP Flutter is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
MCP Flutter Alternatives — Similar Browser Automation Servers
Looking for alternatives to MCP Flutter? Here are other popular browser automation servers you can use with Claude, Cursor, and VS Code.
Chrome DevTools MCP
★ 40.6kAI-powered Chrome automation server with natural language element detection. Control Chrome browser through MCP protocol for testing, debugging, and performance analysis. Features 91% accuracy in element location, works with free AI models, and suppo
UI TARS Desktop
★ 34.9k📇 🏠 - Browser automation capabilities using Puppeteer, both support local and remote browser connection.
Playwright
★ 32.8kA production-ready browser automation server that enables AI assistants to interact with web pages using tools for navigation, element interaction, and data extraction. It features a built-in Inspector UI and robust crash recovery for reliable automa
Page Agent
★ 18.0kJavaScript in-page GUI agent. Control web interfaces with natural language.
Chrome
★ 11.7kAn extension-based MCP server that enables AI assistants to control your browser, leveraging existing sessions and login states for automation and content analysis. It provides over 20 tools for semantic tab search, interactive element manipulation,
LAMDA
★ 7.8kThe most powerful Android RPA agent framework, next generation mobile automation.
Browse More Browser Automation MCP Servers
Explore all browser automation servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up MCP Flutter in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use MCP Flutter?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.