Xcodebuildmcp, from Getsentry, is an MCP server built with Node.js 18+ that provides AI coding agents with direct, programmatic access to Xcode projects, simulators, and physical devices on iOS and macOS. It enables workflows involving xcodebuild and device operations. Distributed via Homebrew or npm, it integrates with various AI clients, allowing advanced automation for Swift development and UI interactions. This tool is designed to bridge AI capabilities with Apple's development ecosystem, enhancing productivity for diverse development teams.

What does Xcodebuildmcp connect to?

Xcodebuildmcp connects directly to Xcode projects, simulators, and physical devices across both iOS and macOS platforms. It offers AI coding agents programmatic access to essential xcodebuild commands, simulator functions, and device workflows. This connection allows AI agents to interact with the Apple development environment, performing tasks that typically require manual Xcode IDE interaction or command-line operations. Its core function is to serve as a bridge, enabling AI systems to operate within the Xcode ecosystem.

Core features and tools

This MCP server exposes up to 82 tools organized into 15 workflow groups. Users can list all available tools using "xcodebuildmcp tools". Notable callable tools include simulator/build, simulator/build-and-run, simulator/test, and simulator/screenshot for simulator management. For debugging, it offers debugging/attach and debugging/breakpoint. UI automation features include ui-automation/tap and ui-automation/swipe. Additionally, when the Xcode IDE bridge is active, tools like xcode_ide_list_tools and xcode_ide_call_tool are accessible.

Who is it for?

Xcodebuildmcp targets AI coding agents, web developers, backend engineers, and data scientists prototyping iOS applications. It also serves professional production teams and organizations seeking enterprise AI integration. The tool supports scenarios where AI agents need to autonomously manage Swift code, compile, debug, and interact with user interfaces. It simplifies iOS development tasks for those less familiar with the Apple ecosystem. Skip this if your development workflow does not involve AI agents interacting with Xcode projects or simulators.

How to get started

No credentials are required for Xcodebuildmcp itself, and the server supports the stdio transport protocol for communication with compatible AI clients. Project-level configuration files, such as ".cursor/mcp.json" or "~/.codex/config.toml", are used to define the server command and arguments, while session defaults can be set via ".xcodebuildmcp/config.yaml" or environment variables.

Use cases

• Autonomous build-debug-fix cycles where AI agents write Swift code, compile, parse errors, fix them, rebuild, launch in a simulator, and interact with the UI.
• iOS prototyping for non-iOS developers by handling xcodebuild and simulator operations.
• UI automation and testing, allowing AI agents to interact with simulator UI elements, capture screenshots, and automate UI testing workflows.

Xcodebuildmcp provides a critical link for AI agents to interact directly with the Apple development ecosystem. It offers a collection of tools for managing Xcode projects, simulators, and devices, supporting advanced automation for Swift development. Its design caters to AI-driven workflows, making iOS and macOS development more accessible to various technical backgrounds. This tool is valuable for integrating AI into the software development lifecycle.