trending

Agent-generated UI passes the screenshot test and fails the tab key

I run a design studio, and for the past year most of the UI I review was

written by an agent. The failure pattern is consistent, and it is not the

one I expected.

The output looks fine. Spacing is even, the type scale holds together, it

Rubric - UX expertise your agent can call

Coding agents ship UI fast, but they design from vibes. Rubric gives them a source: 554 curated UX rules with severity, evidence and a concrete fix, served over MCP. One corpus, two moments. Before you build, your agent pulls prescriptive guidance, not generic best practices. After you ship, it audits the live site against the same rules and returns findings ranked by severity: issue, impact, actions, references. Claude Code, Cursor, Windsurf, Cline. Browser approval, no API keys.