GlanzAI Technical Documentation

    Technical Architecture & Measurement Specifications

    This documentation explains how GlanzAI assesses website readiness and summarizes stored AI-response evidence. It also explains measurement scope, limitations and the correct interpretation of results.

    1. What GlanzAI Currently Does

    GlanzAI is an automated diagnostic and measurement platform that audits website technical signals and queries connected Large Language Model (LLM) providers to evaluate whether a brand is recognized, cited, or recommended in structured prompt scenarios.

    Layer 1: Website Readiness Audit

    Inspects your domain’s live HTTP response headers, /llms.txt file availability, JSON-LD structured data entity graph, robots.txt crawler permissions, and Google Lighthouse performance metrics.

    Layer 2: Multi-Model AI Measurement

    Dispatches standardized buyer-intent, category-comparison, and ecosystem-integration prompt profiles across configured LLM provider endpoints, persisting raw response text and execution latency immutably in Cloudflare D1.

    Evidence-Based Integrity Principle:All metrics in customer reports derive directly from stored raw observation records. If a prompt attempt fails or is unattempted, it is excluded from the denominator. No synthetic metrics, placeholder scores, or invented sentiment values are ever generated.