This project provides an MCP (Model Context Protocol) server for seamless AI assistant integration with Xcode. It empowers AI agents to manage Xcode projects, simulators, builds, tests, CocoaPods, and Swift Package Manager dependencies, offering enhanced error handling and support for diverse project types.

How It Works

The server acts as a bridge, translating AI assistant requests into native Xcode commands. It employs a modular architecture with distinct components for path management, project handling, and tool execution. Key features include robust path validation for security, intelligent project type detection (xcodeproj, xcworkspace, Package.swift), and a tool registry for organizing various Xcode operations. Requests are validated, routed to appropriate tool handlers, and executed, with detailed error reporting for troubleshooting.

Quick Start & Requirements

• Installation: Recommended: ./setup.sh (automates verification, installation, and configuration). Manual: git clone https://github.com/r-huijts/xcode-mcp-server.git, cd xcode-mcp-server, npm install, npm run build, cp .env.example .env, edit .env.
• Prerequisites: macOS, Xcode 14.0+, Node.js 16+, npm/yarn, Swift 5.5+ (for SPM). CocoaPods is optional.
• Setup Time: Automated setup is quick; manual setup involves dependency installation.
• Links: Tools Overview, User Guide.

Highlighted Details

• Supports standard Xcode projects, workspaces, and Swift Package Manager projects.
• Integrates with CocoaPods for dependency management.
• Provides extensive iOS simulator control (booting, app installation, screenshots).
• Offers Xcode utilities like asset catalog compilation and archive validation.

Maintenance & Community

• Contributions are welcome via Pull Requests.
• Development guidelines are provided for adding new tools and features.

Licensing & Compatibility

• MIT License. Permissive for commercial use and closed-source linking.

Limitations & Caveats

The project requires macOS and a recent version of Xcode. While it supports various project types, specific edge cases or highly customized build configurations might require manual adjustments or troubleshooting.