Evaluate screenshots and parse the document object model to locate and interact with page elements reliably. Use this skill whenever the user mentions browser automation, dom parsing, screenshot navigation, element selector, ui interaction, web scraping, headless browser, or is working with Playwright, Puppeteer, Claude in Chrome, even if they never say "browser agent navigation" explicitly. Operates entirely on context the agent already has. Do not use it for unrelated application feature work or general coding questions.
