Algorithmic Art (anthropics) vs Web Artifacts Builder (anthropics)
Algorithmic Art (anthropics) and Web Artifacts Builder (anthropics) are both Visual skills, so an agent choosing between them is matching on descriptions that overlap. Here is where they actually diverge.
Algorithmic Art
anthropics/skills
Art that is generated rather than drawn: flow fields, particle systems, seeded randomness you can re-roll and tune. Code is the medium.
2 scenarios in the bank answer to it
Web Artifacts Builder
anthropics/skills
Ambitious claude.ai artifacts with many moving parts: React, Tailwind, component libraries, real state and routing rather than a single static page.
8 scenarios in the bank answer to it
What is the difference between Algorithmic Art (anthropics) and Web Artifacts Builder (anthropics)?
- Algorithmic Art (anthropics)
- Art that is generated rather than drawn: flow fields, particle systems, seeded randomness you can re-roll and tune. Code is the medium.
- Web Artifacts Builder (anthropics)
- Ambitious claude.ai artifacts with many moving parts: React, Tailwind, component libraries, real state and routing rather than a single static page.
Should I use Algorithmic Art or Web Artifacts Builder?
The clearest answer is a situation each one is unambiguously right for. Both of these are drawn from the game's question bank.
Reach for Algorithmic Art when
You want a header image for your blog that is different on every page load, generated from a seed you can re-roll until you like one.
canvas-design is the close call, because it also produces images. The difference is authored versus generated: 'different every load, from a seed you can re-roll' describes a system that makes images, not one composed image. That is algorithmic-art. Ask for exactly one poster and canvas-design wins instead.
Reach for Web Artifacts Builder when
You're building a small claude.ai artifact that needs tabs, a search box, saved state, and three views of the same data.
Multi-component, stateful, routed artifacts are web-artifacts-builder's stated territory (React + Tailwind + shadcn/ui). Simple one-file artifacts don't need it — tabs, state, and views do.
What they have in common
Both are filed under Visual, the axis along which they collide. That shared ground is what makes an agent pick between them on description alone — and what makes it pick wrong.
Nearby comparisons
Reading the difference is not the same as spotting it at speed. That is the game.
Today's session