!!! This Project is in Active Development, So May Have Unexpected Issues
MCPXcode is an open-source implementation of the Model Context Protocol (MCP) for Xcode. It enables seamless integration between Xcode and AI assistants by providing a structured protocol for context exchange and tool execution within the Xcode environment.
Following the MCP specification, this project creates a bridge between Xcode's development environment and AI tools, allowing for enhanced developer workflows through contextual understanding and programmable interactions. It wraps common command-line tools (xcrun
, xctrace
) and leverages macOS accessibility features to enable AI-assisted automation of Xcode operations.
# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
# Clone the repository
git clone https://github.com/everettjf/MCPXcode.git
cd MCPXcode
source .venv/bin/activate
uv add "mcp[cli]"
{
"mcpServers": {
"MCPXcode": {
"command": "uv", # may use full path
"args": [
"--directory",
"<Path>/MCPXcode", # change to your full path
"run",
"main.py"
]
}
}
}
xcrun Tool Extensions
xctrace Tool Extensions
Accessibility Tool Extensions
MCP HTTP Server
MCP Client Libraries
Tool | Description | Example Prompt |
---|---|---|
list_devices | List all available simulator devices with detailed information | "Show me all available iOS simulator devices and their status." |
boot_device | Boot a specific simulator device identified by its UDID | "Boot the iPhone 14 Pro simulator with UDID A1B2C3D4-E5F6-7890-1234-567890ABCDEF." |
shutdown_device | Shutdown a running simulator device | "Shutdown the currently running iPhone simulator with UDID A1B2C3D4-E5F6-7890-1234-567890ABCDEF." |
install_app | Install an application on a simulator device | "Install the app at path /Users/username/MyApp.app on the iPhone 14 simulator." |
launch_app | Launch an installed application on a simulator device | "Launch the app with bundle ID com.example.myapp on the iPhone 14 simulator." |
Tool | Description | Example Prompt |
---|---|---|
get_sdk_path | Get the path of the current SDK | "What is the path to the current iOS SDK?" |
get_sdk_version | Get the version of the current SDK | "What version of the iOS SDK am I using?" |
get_sdk_platform_path | Get the platform path of the current SDK | "Where are the iOS platform files located?" |
find_developer_tool | Find the path of a specific developer tool | "Where is the 'lldb' binary located in the Xcode toolchain?" |
run_tool_with_sdk | Run a developer tool with a specific SDK | "Run the 'clang' tool with the iOS SDK to compile my source file." |
Tool | Description | Example Prompt |
---|---|---|
xcrun_list_sdks | List all available SDKs with detailed information | "Show me all available SDKs I can build against." |
xcrun_list_schemes | List all schemes in a specified Xcode project | "What are all the build schemes in my MyApp.xcodeproj project?" |
xcrun_build | Build an Xcode project with specified scheme and configuration | "Build my MyApp.xcodeproj project using the Debug configuration for the iPhone simulator." |
xcrun_validate_app | Validate an app before submission to App Store | "Validate my MyApp.ipa file before submitting it to the App Store." |
xcrun_upload_app | Upload an app to App Store Connect | "Upload my validated MyApp.ipa file to App Store Connect for review." |
xcrun_swift_symbols | Extract Swift symbols from a binary file | "Extract and show me all Swift symbols from my app's binary." |
xcrun_otool_headers | Show Mach-O headers of a binary file | "Show me the Mach-O headers of my app's binary file." |
xcrun_otool_libraries | Show linked libraries of a binary file | "What libraries is my app's binary linked against?" |
xcrun_nm_symbols | Show symbols in a binary file | "Show me all symbols in my app's binary file." |
Tool | Description | Example Prompt |
---|---|---|
xctrace_devices | List available devices for tracing | "Show me all devices I can run performance traces on." |
xctrace_templates | List available templates for tracing | "What trace templates are available for performance profiling?" |
xctrace_record | Record app performance using a specified template | "Record a performance trace of my app using the Time Profiler template." |
xctrace_record_advanced | Record app performance with advanced options | "Record a performance trace with custom template options and set the environment variable DEBUG=1." |
xctrace_attach_process | Attach to a running process for tracing | "Attach a performance tracer to my running app with process ID 12345." |
xctrace_export | Export trace data to a specified format | "Export my trace.trace file to JSON format for analysis." |
xctrace_diagnose | Diagnose a trace archive for issues | "Analyze my trace file for any performance issues or problems." |
xctrace_document | Generate documentation for a template | "Create documentation for the Allocations template so I can understand its metrics." |
xctrace_analyze | Analyze a trace file and generate performance reports | "Generate performance reports from my time profiler trace file." |
xctrace_compare | Compare two trace files and generate a comparison report | "Compare the performance between trace files from version 1.0 and version 2.0 of my app." |
This project is licensed under the MIT License - see the LICENSE file for details.
Contributions are welcome! Please feel free to submit a Pull Request.
{
"mcpServers": {
"MCPXcode": {
"env": {},
"args": [
"--directory",
"<Path>/MCPXcode",
"run",
"main.py"
],
"command": "uv"
}
}
}
Seamless access to top MCP servers powering the future of AI integration.