One endpoint. every URL.
One HTTP call, Bearer auth, AIDocument JSON out. Same shape every page on the web resolves to. Pick your starting point below: agent developers head to the quickstart, site owners head to the AI Visitors worker.
Quickstart
Get an API key, POST /v1/aidocument with a URL, parse the response. curl, Python, and Node samples that work as-is. 5 minutes from signup to your first AIDocument.
Read the quickstartSite-owner worker
Optional worker for verified domain owners. Reports AI agent visits from your CDN edge to Lyrenth (Cloudflare and Vercel templates), so Lyrenth sees the full picture of who's reading your site, not just agents that route through our API.
Install a workerDeeper reading
The contract surfaces below are how an AI agent reads Lyrenth without an SDK. Useful both for agents calling our API and for any third party building against the same shape.
AIDocument shape
The canonical v2 JSON envelope every /v1/aidocument response returns. Stable contract, forever.
JSON Schema (draft-07)
Machine-readable contract for the v2 envelope. Validate responses, generate types, point editors at it.
Short markdown index
The llmstxt.org convention. Five lines telling an agent where to read more.
Long-form pitch
Full architecture, every endpoint, the canonical document contract. One fetch, no SDK.
JSON manifest
Capabilities, endpoints, billing model, bot identification. CORS-permissive.
Crawler identification
User-Agent string, robots.txt rules, ownership verification. Site operators land here from access logs.
Pricing model
Per-request billing, what counts toward quota, free tier limits.
Get a key, resolve a URL.
The fastest way to learn the shape is to call it. Free tier ships with 2,000 AIDocuments / month, no credit card.