Same brief. Same model. Only the skill changes.
SkillArena compares UI skills for AI coding agents. Every page here was generated by one model from one product brief. The only thing that differs between rows is the skill added to the system prompt, and the first row uses no skill at all as a control.
Every skill, every brief.
One row per skill, control group first. One column per brief. Select a skill to see all of its runs on skillarena.dev.
Loading the matrix
Fetching the latest results from skillarena.dev.
A controlled experiment.
An honest side-by-side, so you can judge what a skill actually does to the output.
Hold everything constant. Change one variable. Look at the pages.
- 01
Same brief, same model
Each page is generated from the same product brief by the same model. The only change between rows is the skill in the system prompt.
- 02
Three runs per pair
Every skill and brief pair is run three times, so a single lucky or unlucky generation doesn't decide the result.
- 03
Full results on skillarena.dev
Every skill and thumbnail links to its full results page, with all runs side by side. Start at skillarena.dev.
Get the next round.
New skills, new briefs and new models. We'll email you when results land.
Updates only.
No noise.
Fields: email · role