Engineering checklist: what it takes for an LLM to actually cite your brand

A technical audit conducted on 7 August 2026 tested 12 Brazilian generative engine optimization domains to assess their readiness for AI crawler indexing. Two of the audited domains were blocking AI crawlers like GPTBot with 403 responses while marketing AI visibility services, undermining their own credibility. The audit found that only two domains — alexandrecaramaschi.com and brasilgeo.ai — served both llms-full.txt files and a Model Context Protocol server surface, making them directly queryable by AI agents. Key engineering requirements identified include explicitly listing AI user agents in robots.txt, generating structured data with stable @id identifiers, and publishing llms.txt and llms-full.txt files built from the same source as the sitemap. The findings highlight that technical infrastructure, not content strategy alone, determines whether a brand appears inside AI-generated answers.