What problem does it solve? Manually driving a browser to test web applications, fill forms, or scrape page data is repetitive and error-prone. This Skill provides a command-line interface to Playwright so an AI agent can navigate pages, interact with elements, capture snapshots, and manage browser state without writing a full test harness first. ## Core Features & Use Cases - Page Interaction and Navigation: Open pages, click elements by snapshot refs, fill forms, press keys, drag elements, and move through tabs, history, and iframes. - State and Session Management: Run multiple isolated named browser sessions, persist profiles, and save or restore cookies, localStorage, sessionStorage, and full storage state files. - Network Control and Debugging: Mock or block requests with route rules, inspect console and network logs, record traces and videos, and execute arbitrary Playwright code via run-code. - Test Generation: Every CLI action emits corresponding Playwright TypeScript code that can be pasted directly into test files. - Use Case: Log into a web app, save the authenticated storage state to auth.json, then reuse that state in later sessions to skip the login flow while scraping dashboard data. ## Quick Start Ask the agent to open a website with playwright-cli, take a snapshot to list interactive elements, then fill the login form and click the submit button.
