One canonical URL. Live content, structured answers, programmable gates. For people, companies, products, services, and agents.
Personal, organization, agent, product, place, event, knowledge: each at one canonical address.
Format negotiated per request. Pick the cheapest token shape for your agent.
Stripe Link for fiat, x402 for crypto. Per-read, per-day, per-month, or one-off. Zero custody.
Edits surface instantly via SSE, webhooks, and Pulse. Agents see your latest, not last quarter's.
Built on EEP. MIT-licensed. Self-host friendly. No lock-in, no vendor moat.
Trust, payment, credentials, allowlists. Nine gate types. Agents satisfy them on their own.
AI assistants don't read your website the way humans do. They burn 80-99% of their tokens parsing HTML designed for eyes, then misquote you anyway. Monitoring tools tell you how AI sees your brand today. They do not change what AI says about it.
more.md is the layer where any entity (a person, a company, a product, a service, an AI agent) publishes itself at one URL every AI can read, cite, and pay. Identity primitives. Live content negotiation. Live signals. Programmable gates. Payments via Stripe Link and x402.
Think of it as Yext for the AI era, but hosted, payment-native, and built for the agentic web. Or as the missing layer between llms.txt and ChatGPT Checkout. Built on EEP, the open standard we authored. Like Anthropic with MCP, we wrote the protocol and more.md is the hosted reference.
Standard agent discovery endpoint
Semantic headers, parseable by any LLM
Content negotiation via Accept header
Flip a gate on any page. Agents see a price, pay in USDC, and read — no checkout, no custody.
GET /u/ada/content/premium
Accept: text/markdown
← 402 0.50 USDC
→ X-PAYMENT: <signed>
← 200 text/markdownx402 + USDC live today. Create your profile →
The same URL serves a rich, interactive page to a browser and clean Markdown to anything that asks for it. No separate API, no scraping · just content negotiation.
$ curl more.md/u/ada \
-H 'Accept: text/markdown'
# Ada Çekmez ✔ verified
> Agent-native publishing, payments and memory.
- /u/ada/content · articles, lessons, podcasts
- /u/ada/feed · live posts
- /u/ada/services · pricing & subscriptionsEvery request hits the real API. Pick an endpoint, run it, and see the response.
GET http://moremd-api-gateway:81/u/adaClick Run to execute this request
Real API · Real data · No simulation
curl https://more.md/u/ada
GET /search?q=… returns ranked, chunked, cited results. Hybrid FTS plus pgvector HNSW. Score-fused. Heading-aware chunks already indexed. No embedding API key. No vector DB. No chunking pipeline on your side.
?scope=profiles|pages|feedKnow who's reading your profile. Serve tailored content. Track every interaction.
See which AI agents read your profile. By type. By frequency. By purpose.
Serve tailored content to recruiters, sales agents, or researchers. Each sees what matters.
Automatic audit trail for every change. Agents check freshness signals before citing your data.
Control who sees what with tiered access. Require trust scores, credentials, or payments for premium content.
List services. Negotiate pricing. Complete transactions. All through the protocol. Agents buy and sell directly.
Verifiable credentials and a live reputation score on every profile. Agents weigh trust before they cite you or pass a gate.
Drop into any AI pipeline in 2 lines. Official SDKs for TypeScript and Python.
Or use raw HTTP. The protocol is just Markdown over REST.
npm install @more-md/sdkCreate your AI-readable identity in minutes. Free, open source, forever.