flutter ilot
flutter, mcp, modular, code-generator, ai-assistant, .net, devtools, code-review, test-generator, refactor
claude mcp add --transport stdio alperduzgun-flutter-mcpilot-server dotnet run
How to use
Flutter MCP Server is a project-aware, modular backend assistant designed to work with Flutter projects. It provides a suite of specialized MCP commands across environment management, code generation, analysis, testing, documentation, packaging, and file-system operations. By leveraging its modular handlers, it can generate Dart classes, Cubits, API services, and theme modules; migrate navigation systems to GoRouter; perform code reviews and complexity analysis; generate unit tests using blocTest patterns; search Flutter docs and pub.dev packages; and safely perform file operations with path traversal protection. The server speaks the MCP protocol over REST, JSON-RPC 2.0, and MCP endpoints, enabling clients to interact through standard interfaces and tooling. When you run it, the server maintains context about your project, learns from past decisions, and offers explanations for why certain architectural choices or code patterns are suggested, not just what to do next.
How to install
Prerequisites:
- .NET 7+ installed on your machine
- Git installed
- A Flutter project (optional, for testing with project context)
- Clone the repository
- git clone https://github.com/your-org/flutter-mcp-server.git
- cd flutter-mcp-server
- Restore and run the server
- dotnet restore
- dotnet run
- Verify the server is running
- Access the MCP API endpoints via the configured port (default typical port of the project) and test a few commands like migrateNavigationSystem or generateDartClass using an MCP client.
- Optional: configure environment variables (if needed by your deployment)
- Create an environment file or set variables in your hosting environment as required by your deployment scenario.
Additional notes
Tips and considerations:
- The server uses a modular CommandHandler architecture with dedicated handlers for Environment, CodeGeneration, Analysis, Testing, Documentation, Package, and FileSystem tasks. You can extend it by adding new handlers under the Handlers/ directory.
- Enable dry-run mode in your MCP client to preview changes before they are written. This helps in safe experimentation.
- Review the project log at project_log.txt to understand command telemetry and performance insights.
- Ensure your Flutter project structure is compatible with the generated outputs (e.g., Dart class templates, Cubit boilerplate, and API service patterns).
- When migrating navigation or refactoring architecture, verify that the generated code adheres to recommended Flutter best practices and the GoRouter migration patterns supported by the server.
- If you customize MCP capabilities, maintain the mcp-capabilities.json template in Config to reflect available commands and categories.
Related MCP Servers
mssql
MSSQL Server MCP implementation written in C#
McpDotNet.Extensions.SemanticKernel
Microsoft SemanticKernel integration for the Model Context Protocol (MCP). Enables seamless use of MCP tools as AI functions.
DatabaseMcpServer
MCP server from ttcc666/DatabaseMcpServer
mcp-dataverse
MCP Server for querying Dataverse using SQL
xperience-community
ASP.NET Core MCP server for Xperience by Kentico projects
console-to-http
Example of converting a stdio MCP server to HTTP using ModelContextProtocol.AspNetCore