MCP for car dealers: what it is and why yours should be live already
Model Context Protocol is how AI agents will check your inventory, hours, and service availability without scraping your site. PromptGraph runs it per rooftop today.
Every AI vendor in automotive has an MCP slide now. Here is the part they leave off the slide: MCP is an open protocol that Anthropic published in late 2024, and it is not hard to deploy. What is hard is having a per-store server with real inventory behind it, running, today. PromptGraph does. This page explains what that means for a dealer in plain English.
The problem MCP solves
When a shopper asks ChatGPT "does anyone near me have a white F-150 Lariat under $60k," the model has three ways to answer. It can guess from training data, which is months old. It can search the web and read whatever pages it can parse, which for most dealer sites means fighting through a JavaScript inventory grid. Or it can call a tool that returns the answer as structured data.
MCP is the standard for that third option. A dealer exposes a small server with a handful of tools: search inventory, get vehicle details, get store hours, check service availability, get directions. An AI agent connects, asks, and gets JSON back. No scraping, no guessing, no ten-second-old page load that missed the price change.
What a dealer MCP server actually exposes
PromptGraph's per-rooftop servers expose the store the way an agent needs it:
- Inventory search by year, make, model, trim, price range, body style, and drivetrain, returning live units with stock number, VIN, price, and a link to the VDP.
- Vehicle detail for a single VIN: photos, options, window sticker data where available.
- Store profile: name, address, phone, hours by department, brands sold, and the Google Business Profile link so the agent can match entities.
- Service: hours, offered services, and a scheduling link.
Each server publishes a manifest at a well-known path so an agent can discover it without being told where to look.
Why this matters more than schema
Schema markup describes your store. MCP lets an agent act on it. Across the 2,699 franchise sites we could score in September 2026, 96% had AutoDealer schema and 12 had an MCP endpoint. That is 0.4%. Schema is table stakes. The agent layer is where the whitespace is, and it carries 55 of the 100 points in our readiness score for exactly that reason.
What it costs a dealer
Nothing beyond the PromptGraph subscription. The server runs on our infrastructure, reads your inventory feed, and is registered against your domain. No IT project, no platform change, no waiting for your website vendor's roadmap.
What it does not do yet
No consumer AI assistant auto-discovers every dealer MCP server on the internet today. Adoption is moving engine by engine and integration by integration. What is true today is that agents built on Claude, on OpenAI's tooling, and on a growing list of shopping assistants can connect to a published MCP server, and the stores that are live get the traffic when each engine flips the switch. The stores that wait get a slide.
Do I need to change my website platform to run MCP?
No. The server runs separately from your website and reads your inventory feed. Dealer.com, DealerOn, Dealer Inspire, and OEM-mandated sites all work.
Is MCP the same as llms.txt?
No. llms.txt is a text file that tells a language model what your site is about. MCP is a live server an agent can query for current data. You want both. They are two different checks in the readiness score.
Can I see a live dealer MCP server?
Yes. Ask for a demo at Gray@PromptGraph.ai and we will connect an agent to a client rooftop while you watch.
See where your store stands today.
PromptGraph runs the readiness checks and the prompt-based visibility test on your rooftop, then deploys the fixes. Free scan first, no card.
Run the free AI visibility scan