Skip to the content.

auto-geo

Publishing engine for GEO resource pages — the pages large language models cite. Built by Shadow, a media research lab building the next generation of AI-powered media intelligence and communications technology, in partnership with the teams that put OpenAI, TikTok, Meta, Amazon, and Lovable on the map.

auto-geo is a typed publishing primitive for Generative Engine Optimization resource pages — pages whose structure, density, and citation signals are engineered to be quoted by AI search engines (ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini) when they answer a user’s question.

It validates an incoming payload against the full GEO architecture at the publish boundary, derives Schema.org JSON-LD automatically, and ships HTTP adapters for Next.js and Hono plus storage adapters for Vercel KV / Upstash Redis, Supabase, and in-memory testing.

Read this first

Reference

LLM-friendly

This site emits an llms.txt following the llmstxt.org convention, plus an llms-full.txt that inlines the README + concept + SOP + architecture + validation + storage-adapter docs into a single ingestible bundle. Every page in this site also advertises both files via <link rel="alternate" type="text/plain"> in the <head> so well-behaved crawlers and AI agents discover them automatically.

Examples

Working minimal apps for the most common backends — each boots locally with pnpm install plus the framework’s dev script: