Lighthouse 13.3 introduces a new 'Agentic Browsing' category that audits how well websites work with AI agents. The category is still marked as under development and includes four checks: accessibility tree well-formedness, layout shift detection (CLS), WebMCP implementation validation, and llms.txt file compliance. WebMCP is a browser-native API allowing websites to expose commands to AI agents via annotated HTML forms or the programmatic navigator.modelContext.registerTool API. The llms.txt audit checks for proper formatting of the file. Notably, sites without any AI-specific features can still score well — example.com scores 2/2. The new category can be tested via the Lighthouse CLI (v13.3+) or third-party tools, as PageSpeed Insights and Chrome DevTools still use an older version.