What problem does it solve? Verifying what a browser actually renders, reproducing UI bugs, and capturing visual evidence often requires manual clicking through pages. This Skill drives a real browser from the terminal so agents can navigate, fill forms, snapshot pages, and capture screenshots without writing test files. ## Core Features & Use Cases - CLI-first browser automation: Open pages, snapshot the DOM to get stable element refs, click, type, fill forms, and navigate using playwright-cli commands. - Evidence capture: Take screenshots, PDFs, console logs, network logs, and traces to document UI behavior or post-deploy verification. - Session isolation: Use named sessions to keep separate workflows isolated and avoid state conflicts. - Use Case: After a deployment, open the production site with the wrapper script, snapshot the page, click through a key flow, and capture a screenshot proving the page rendered and the interaction still works. ## Quick Start Use the playwright skill to open https://example.com in a browser, snapshot the page, and take a screenshot to confirm it renders correctly.