AgentSwift is a new open-source macOS app from developer hpennington that runs an autonomous AI coding agent for Apple platform development. You describe what you want built, and it uses Anthropic's Claude models to discover your Xcode project, write code, compile it, launch it in a simulator, and verify the behavior through UI automation. The whole cycle runs without you touching Xcode, a topic that sparked a lively discussion when the app launched. The tool works through a multi-phase agentic loop: discover, implement, build, launch and validate, then archive. It depends on two command-line utilities, xcodebuildmcp and openspec, which handle build logic and spec tracking across sessions. As developers increasingly rely on the Model Context Protocol to connect AI to external tools, understanding the broader adoption landscape can provide helpful context.
AgentSwift gives Claude the keys to your Xcode projects
AgentSwift is a native macOS app that runs an autonomous AI coding agent for Apple platform development. It uses Claude to discover Xcode projects, implement changes, build, run, and validate iOS applications without user intervention.