screenwrightlisted

# screenwright > **The wedge is the eyes.** Generating UI is solved; *verifying the render against a > brand + accessibility bar, in a loop* is not. screenwright paints one self-contained > HTML surface, then renders → screenshots → audits → fixes until it passes, and hands the > result back. It does the painting and verification; it does **not** ship a runtime or > convert to a framework — that's the host's job. Agents write frontend blind: the markup looks right but misses the small obvious things — a 4px misalignment, a missing focus ring, text that overflows on a long name, a mobile layout that never got built. screenwright gives the agent eyes. **Two stacked gates, asymmetric:** - **axe a11y** — machine, deterministic, **always blocks**. The defensible core. - **Visual fidelity** — judgment from the screenshot. **Blocks only when a reference image exists**; advisory otherwise (no reference = no ground truth to gate on). --- ## Phase 0 — Capability gate (run first) screenwright renders through the **Playwright MCP**. It is a **hard dependency** — gate on its presence, not on which agent you are (capability-gated, not vendor-gated). - If Playwright MCP tools (`browser_navigate`, `browser_run_code_unsafe`, `browser_take_screenshot`) are **not** available → **stop**. Tell the user screenwright needs the Playwright MCP, give install instructions (`claude mcp add playwright npx '@playwright/mcp@latest'` or the host's equivalent), and do not attempt a degraded