A developer details how he shipped Context, a native macOS app that was almost 100% built using Claude Code
I recently shipped Context, a native macOS app for debugging MCP servers. The goal was to build a useful developer tool that feels at home on the platform, powered by Apple's SwiftUI framework. Bluesky: @johnspurlock.com . X: @iannuttall , @kirchnerus , @ceccon_me , @gergelyorosz , and @indragie . Forums: Hacker News Bluesky: John Spurlock / @johnspurlock.com : 'The most exciting thing about this entire journey for me is not the app I built, but that I am now able to scratch my coding itch and ship polished side projects again. It's like I found an extra 5 hours every day, and all it cost me was $200 a month.' — www.indragie.com/blog/i-shipp... X: Ian Nuttall / @iannuttall : This is a great read on using AI to build a macOS app and principles apply to all apps and coding tools Michael Kirchner / @kirchnerus : I read many developer blogs, but this is one of the most useful and direct I've found in months. Thanks for the great article, @indragie! Claude Code + XcodeBuildMCP = pure bliss for dev experience. I've long awaited a smooth AI workflow for iOS, and this is it. Francesco / @ceccon_me : Very good read. I've also found that priming a new session with the content of key files is super helpful to get good results. Never tried ultrathink, usually I don't leave much room for choice Gergely Orosz / @gergelyorosz : This is a very good read for any developer: *especially* if you have not yet used CLI agents. These tools can and do speed up development of greenfield projects greatly, and make it more pleasant and thorough. Eg using Claude to generate mock data or build a much cleaner UI: Indragie Karunaratne / @indragie : I haven't written a blog post in 11 years, but the recent experience of building a complete macOS app with AI has me so optimistic about the future of software development that I had to share more about the process. [image] Forums: Hacker News : Building a Mac app with Claude code
Context & Ripple Effects
Claude's capabilities had already expanded from running code and analyzing files to creating prompt-built interactive apps. Context applies that progression to a conventional native Mac developer tool rather than a browser-based prototype.
The project arrives as Claude Code is being presented as capable of writing much of its own code, while Apple has opened Xcode 26 to external model providers. That makes this a concrete example of AI-assisted development crossing into platform-native software.
First-order effects
- Indragie Karunaratne has shipped a SwiftUI-based macOS utility for debugging MCP servers, using Claude Code for nearly all of its implementation.
- The account provides a practitioner-level proof point for Claude Code's high-autonomy coding workflow on a polished side project, not merely for snippets or experiments.
Second-order effects
- Developers building MCP-connected tools gain a native debugging option, while AI coding products face a higher bar: helping users complete and ship platform-specific applications.
- Apple's move to let developers bring other models into Xcode makes projects such as Context more relevant to the model-flexible Xcode workflow rather than to a single assistant or IDE.
Third-order effects
- If comparable projects can be repeatedly shipped with limited manual implementation, small-team and solo software development may shift toward faster experimentation and more specialized developer utilities.
- The limiting differentiator moves toward problem selection, product judgment, testing, and integration context—not simply the ability to generate code—especially for tools that sit around MCP-based workflows.
The trend: AI coding assistants are evolving from code-completion features into agentic work surfaces that can support end-to-end delivery of narrowly scoped native software.