Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-03-06 · source-backed
— "The defining characteristic of a coding agent is that it can execute the code it writes." Never assume LLM-generated code works without verification. Patterns for python -c edge case testing, /tmp demo files, browser automation with Playwright/Rodney. Red/green TDD: when agents find issues, add permanent test coverage. (simonwillison.net)
Each link below shares sources, entities, or timing with this story.
- Source: simonwillison.net - Date: 2026-02-10 Two tools addressing the core verification problem: how do you confirm what a coding agent built? Showboat creates executable markdown documents mixing commentary, code, and captured output. Rodney provides browser automation for...
Simon Willison launched a multi-chapter guide collecting practical patterns for working with coding agents. Chapters include "Writing code is cheap now," "First run the tests," "Red/green TDD," "Hoard things you know how to do," and the new "Interactive Explanations" chapter f...
After 20+ years maintaining Paint.NET, Rick Brewster concluded WINE's Direct2D would never be complete enough for what he needed, so the app now carries its own from-scratch reverse-engineered Direct2D implementation. He puts it at 180,000 lines against 700,000 for the rest of...
Simon Willison spent a while taking ChatGPT Work apart and published the map on August 30. Work splits into Work Cloud and Work Local, the latter being the renamed Codex desktop app, at $20/month and up since July 9. He enumerates six capabilities Work has that Chat doesn't, a...
Simon Willison added a shot-scraper video command that takes a storyboard.yml and uses Playwright to record a walkthrough of a web app, and his demo storyboard was authored entirely by GPT-5.5 xhigh in Codex Desktop. This is a genuinely useful primitive: agent-generated PR dem...
I've been skeptical of "just have AI rewrite your codebase" advice because most of it comes from people who haven't actually done it in production. This week, three independent teams published results that changed my mind, with a very specific caveat. Reco.ai rewrote JSONata (...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.