XCF Xcode MCP Server

XCF Xcode MCP Server is an MCP server that integrates macOS and iOS Xcode workflows directly into AI development environments. Built in 100% Swift with zero external dependencies, it connects Xcode with AI code assistants such as Cursor, Claude Desktop, and VS Code. Swift developers and Apple platform engineers use it to control local Xcode project lifecycle tasks programmatically from AI chat sessions. The server exposes twenty-one MCP tools covering project management, file system inspection, and direct code intelligence. It enables AI agents to query open Xcode projects, trigger builds, execute applications, capture real-time compilation errors, and inspect directories. Additionally, it offers standalone Swift code analysis without requiring an active Xcode build, identifying structural complexity, unused symbols, magic numbers, and style issues. Using fuzzy logic and smart path resolution, the server resolves project files using filenames or partial paths across defined workspace directories.

Category: Developer Tools & Code Intelligence

Tags: ide, ios, macos, swift, xcode

Visit XCF Xcode MCP Server

How to install and configure XCF Xcode MCP Server

1. Download the XCF application and move it to your /Applications folder, or build it locally from source using Xcode. 2. Open the application once to clear the macOS download alert, then click "Press to Quit this XCF Xcode MCP Server". If it does not show, run codesign --force --deep --sign - /Applications/xcf.app. 3. Add the server entry to your MCP configuration file (such as ~/.cursor/mcp.json or ~/Library/Application Support/Claude/claude_desktop_config.json): json { "mcpServers": { "xcf": { "type": "stdio", "command": "Applications/xcf.app/Contents/MacOS/xcf", "args": ["server"] } } } 4. Restart your MCP client or refresh tools to activate the integration.

What you can do with XCF Xcode MCP Server

• Building and running Xcode projects directly from Cursor or Claude Desktop to automate application testing. - Forwarding real-time compilation errors and warnings to your AI assistant to generate immediate bug fixes. - Inspecting project directories, changing folders, and extracting targeted code snippets using fuzzy path matching. - Analyzing Swift source files for high complexity, style issues, and unused symbols without running a full Xcode build. - Managing multiple open Xcode projects by listing open instances and switching active targets through AI tool calls.

Key facts

• https://github.com/CodeFreezeAI/xcf
• Developer Tools & Code Intelligence, DevOps, CI/CD & Version Control
• ide, ios, macos, swift, xcode

Part of MCP Servers

What is XCF Xcode MCP Server?

XCF Xcode MCP Server is an integration server built in Swift that connects Xcode with AI clients like Cursor and Claude. It provides tools for building projects, reading code, running apps, and running Swift static analysis.

Which MCP clients work with XCF Xcode MCP Server?

The server works with standard Model Context Protocol clients that support local stdio communication. It is officially documented for Cursor, Claude Desktop, Claude Code, and VS Code on macOS systems running Xcode.

How do I install XCF Xcode MCP Server?

Download the prebuilt application and move it to /Applications, or build the project from source using Xcode. Approve the app once, quit it, and configure the executable path inside your client mcp.json file under the stdio transport setting.

What can XCF Xcode MCP Server do?

It provides 21 tools covering Xcode automation, directory browsing, file snippet retrieval, and Swift code analysis. Users can compile projects, switch between workspaces, resolve file paths using fuzzy matching, and identify code quality issues without building.

Does XCF Xcode MCP Server require Node.js or Python?

No. The server is built entirely in Swift 6.1 with zero external runtime dependencies. It runs directly as a native macOS executable without requiring JavaScript, TypeScript, Node.js, or Python runtimes.