Create a Cross-Browser Visual Regression Matrix

ROLE Act as a frontend visual-quality engineer. CONTEXT Pages and components: [INTERFACE INVENTORY] Supported browsers and devices: [SUPPORT POLICY] Design references: [DESIGN SOURCE] Screenshot or testing tools: [AVAILABLE TOOLS] OBJECTIVE Build a compact visual regression plan that catches meaningful presentation changes. INSTRUCTIONS Group interfaces by shared layout patterns and select representative coverage. Include high-risk states such as long content, empty results, validation errors, expanded menus, and localized text where applicable. Choose viewport boundaries based on layout behavior rather than arbitrary device lists. Explain how to stabilize fonts, animations, dates, and network-dependent content before screenshots. Define baseline creation, review ownership, acceptable rendering variation, and escalation for unexpected differences. Separate visual checks from functional assertions so screenshots are not mistaken for complete behavior tests. CONSTRAINTS Do not fabricate coverage statistics or claim screenshots were captured. Flag assumptions and tool limitations. Avoid masking large regions simply to make comparisons pass. OUTPUT FORMAT Return a coverage matrix, capture setup, baseline policy, and triage decision tree. QUALITY CHECK Explain which important defects the chosen matrix could still miss.