One MCP server. Every agent protocol.
MCP, A2A, UCP, llms.txt, schema.org — your live catalog readable by every AI agent through open standards. No one-off integrations to build, no closed APIs to maintain.
Open standards. No closed surface.
Every agent reads your catalog the same way — through the protocol it already speaks. We track each spec as it evolves and ship the surface so you don't have to.
Model Context Protocol
A public MCP server at mcp.chatcast.io. Discovery, tool listing, and tool invocation so ChatGPT, Claude and Perplexity can query your catalog, FAQs and policies in real time. No agent-specific integration to maintain.
Agent-to-Agent
Structured handoff between agents — a shopper's research agent passes context to a checkout agent without losing the cart, the discount code or the citation receipt. Built for the multi-agent flows that are emerging now.
Universal Commerce Protocol
An open spec for commerce-native agent capabilities — cart, checkout, inventory, returns. We track the standard as it evolves and expose the surface area you need to participate.
Discovery for the agentic web
A signed llms.txt published at your domain root and refreshed as your catalog changes. The robots.txt of the agentic web — tells agents what to read, what to skip, and where the canonical machine-readable catalog lives.
Structured product graph
Product, Offer, FAQPage, BreadcrumbList, AggregateRating — fully populated and validated. The schema agents actually parse, with the attributes that drive citations.
Common questions.
01What is an MCP server for a Shopify store?
MCP (Model Context Protocol) is the open standard AI assistants use to query external data. A Chatcast MCP server is a public endpoint for your store that lets ChatGPT, Claude, Perplexity and other agents search your live catalog, FAQs and policies in real time instead of relying on stale crawls.
02Do I need a developer to set this up?
No. Chatcast hosts and maintains the server; you enable it from the console. The discovery signals — llms.txt, .well-known/mcp.json, schema.org markup — are published automatically.
03Is it safe to expose my catalog to AI agents?
The server exposes only public storefront data — products, FAQs and the policies you choose to publish. It requires no customer data, and requests are rate-limited per IP per store.
04Which AI agents can connect?
Any MCP-compatible client — ChatGPT, Claude, Cursor and the growing ecosystem of shopping and research agents. One open standard instead of one-off integrations per platform.
05How do agents find my store in the first place?
Chatcast publishes machine-readable discovery signals for your domain: an llms.txt agent guide, a .well-known MCP manifest, and schema.org structured data. Agents that resolve any of them find the endpoint.
Get a free product data quality audit.
Find out which product attributes LLMs are missing on your catalog — and what it's costing you in agentic search visibility. No call required, the report lands in your inbox.