
Your agent can already build a website, a slide deck, an image, even a short video. What it usually lacks is taste. Skillry is a growing library of Agent Skills that supply exactly that — each one carrying a specific visual point of view, so the work your agent ships looks deliberately made instead of generically assembled.
Nothing here is a blind purchase. Before you install a Skill, you can see precisely what it produces, which means you are judging real output rather than trusting a written description. What your agent delivers ends up being something you would actually show to a client, a team, or an audience.
Getting started takes moments. Install from the web, or authorize your browser once and run a single command in your CLI. Then simply invoke a Skill by name inside your agent and receive a finished piece of work: a landing page that looks art-directed, a deck you would genuinely present, an image that stops the scroll, or a short film that moves people.
Today the library covers four kinds of work — web pages, slide decks, images and video — with new Skills added every week. No API key is required, so there is no setup friction between an idea and a polished result.
Skillry is deliberately not a marketplace. There is no community upload and no race to ten thousand listings. The collection stays small and grows one Skill at a time, because that is the entire point: everything inside it has a look worth keeping.
It works with Claude Code, Codex, Cursor and more, fitting directly into the tools you already use. The taste is ours. The work is your agent's.

Subscribe to get the latest Skillry and global AI updates.
Comments
Loading comments…