How audited pages become queryable knowledge twins — and how every twin tool consumes them.
In this platform, a site's digital twin is the structured knowledge base produced by an AEO audit and stored in AeoReport.report_data. It is not a copy of the page — it is the answer-ready distillation: answer blocks, Q&A pairs, entity mappings, question clusters, passage analysis, speakable schema, and the full JSON-LD library.
Every tool that reasons about a site reads from this twin rather than re-scraping, which makes queries instant and grounded in content you have already audited.
AeoReport.report_data under your email + domain.The AEO Knowledge Assistant (/chat) and the backend siteChat function load the twin for a selected domain and answer natural-language questions with citations back to the audited source blocks. Voice input is supported in Chromium browsers.
The Site Twin agent exposes the same twin over WhatsApp and Telegram via connect links (base44.agents.getWhatsAppConnectURL / getTelegramConnectURL). Conversations are managed with the agents SDK and the agent reads the AeoReport entity for grounded answers.
/compare-twins loads two domains' twins side by side — content gaps, entity coverage, question coverage, and schema — to show exactly where a competitor is more citable than you.
The Brand Truth / Hallucination Guard uses your twin as ground truth: it simulates what AI engines say about your brand, breaks the answer into atomic claims, and verifies each claim against your audited content. Our self-audit caught one hallucination — AI engines fabricated a 2023 founding date because no About section exists in our twin.
Our own twin for seotoaeoai.com scored 78/100. The twin is seeded but thin (client-rendered SPA), so it currently answers the brand query well but lacks the depth to win category queries. Remediation: add substantive copy, headings, JSON-LD, and an About section — then re-audit to enrich the twin.