Codex vs Claude Code: Same One-Shot Prompt, Two Websites
TL;DR: I wrote one detailed prompt, gave the exact same prompt to ChatGPT Codex and Claude Code in VS Code, and let each build a shop website in one shot. No planning step, no follow-up fixes to the design. Codex finished first but its preview didn't open at first. Claude took longer and gave me a more interactive site with more images. I liked Claude's version more, so I'm continuing with it.…
The author conducted an experiment to compare the results of using the same one-shot prompt with two AI coding agents, Claude Code and ChatGPT Codex. Both AI agents were given the same detailed prompt to create a website for a shop selling four battery-powered garden tools. The goal was to see how far each AI could go in generating a base website layout, design system, components, and structure without further coding or design input.
The author used VS Code to run the experiment, creating separate project folders for each AI agent to avoid file overlap. The prompt was generated by analyzing a highly detailed website prompt created with Gemini 3.1 Pro. The author then fed this prompt to both AI agents, observing their respective progress, output, and the final results.
Claude Code took longer to finish but produced a more interactive website with more images and sections. ChatGPT Codex, on the other hand, finished first but had a preview issue that required additional steps to resolve. The author ultimately preferred Claude Code's version due to its more engaging design and overall presentation.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
