Simon Willison’s Weblog

Vibing a Non-Trivial Ghostty Feature (via) Mitchell Hashimoto provides a comprehensive answer to the frequent demand for a detailed description of shipping a non-trivial production feature to an existing project using AI-assistance. In this case it's a slick unobtrusive auto-update UI for his Ghostty terminal emulator, written in Swift.

Mitchell shares full transcripts of the 16 coding sessions he carried out using Amp Code across 2 days and around 8 hours of computer time, at a token cost of $15.98.

Amp has the nicest shared transcript feature of any of the coding agent tools, as seen in this example. I'd love to see Claude Code and Codex CLI and Gemini CLI and friends imitate this.

There are plenty of useful tips in here. I like this note about the importance of a cleanup step:

The cleanup step is really important. To cleanup effectively you have to have a pretty good understanding of the code, so this forces me to not blindly accept AI-written code. Subsequently, better organized and documented code helps future agentic sessions perform better.

I sometimes tongue-in-cheek refer to this as the "anti-slop session".

And this on how sometimes you can write manual code in a way that puts the agent the right track:

I spent some time manually restructured the view model. This involved switching to a tagged union rather than the struct with a bunch of optionals. I renamed some types, moved stuff around.

I knew from experience that this small bit of manual work in the middle would set the agents up for success in future sessions for both the frontend and backend. After completing it, I continued with a marathon set of cleanup sessions.

Here's one of those refactoring prompts:

Turn each @macos/Sources/Features/Update/UpdatePopoverView.swift case into a dedicated fileprivate Swift view that takes the typed value as its parameter so that we can remove the guards.

Mitchell advises ending every session with a prompt like this one, asking the agent about any obvious omissions:

Are there any other improvements you can see to be made with the @macos/Sources/Features/Update feature? Don't write any code. Consult the oracle. Consider parts of the code that can also get more unit tests added.

("Consult the oracle" is an Amp-specific pattern for running a task through a more expensive, more capable model.)

Is this all worthwhile? Mitchell thinks so:

Many people on the internet argue whether AI enables you to work faster or not. In this case, I think I shipped this faster than I would have if I had done it all myself, in particular because iterating on minor SwiftUI styling is so tedious and time consuming for me personally and AI does it so well.

I think the faster/slower argument for me personally is missing the thing I like the most: the AI can work for me while I step away to do other things.