Auxy tracks selected citations by using provider-specific parsers across Google, OpenAI, and Anthropic grounded responses to extract source URLs, cited text snippets, and character offsets directly from grounding metadata. These parsed citations are resolved to their canonical destinations, cross-referenced against configured domain ownership rules to flag owned versus competitor sources, and stored in the database alongside their relative ranking and specific brand mentions.