mcp server
marketplace
AI-native used car marketplace. 145K+ vehicles, 4300+ dealers, 13 US states, 20 MCP tools.
Description as published by the maintainer. Source
- version 1.0.0
- slowing
slowing — Registry entry last updated 2026-05-07.
What this server can do
19 functions, named and described by the server itself. Parameter names are shown because they say more about what a function does than its name usually does.
cars_dealer_inventory(limit, offset, dealer_id)- Get all vehicles at a specific dealer. Required: dealer_id.
cars_dealers(zip, city, limit, state)- Search for car dealers by city, state, or ZIP code.
cars_feedback(rating, message, session, agent_id, category, user_happy)- Submit feedback about the Cars Rootz service. We actively read every piece of feedback to improve the service. Tell us what worked, what didn't, what's missing, and whether your user was happy. This helps us build a better car shopping experience for everyone. Required: category, message.
cars_history(vin)- Get the full history of a vehicle by VIN — every dealer it appeared at, price changes over time, days on market, whether it was sold and resurfaced elsewhere. Data with origin. Required: vin.
cars_search(zip, city, make, limit, model, state, offset, year_max, year_min, body_type, fuel_type, price_max, price_min, drivetrain, mileage_max)- Search used car inventory across all dealers. Filter by make, model, year, price, mileage, location, body type, fuel type, drivetrain. Returns matching vehicles with dealer info.
cars_session_add_car(vin, notes, score, session, agent_id)- Add a vehicle to the buyer's shopping session. Include a fit score (1-10) and notes explaining why you recommend it. Required: session, vin.
cars_session_compare(session)- Get a structured comparison of all vehicles tracked in a session — specs, pricing, market context side by side. Great for helping the buyer decide. Required: session.
cars_session_create(zip, name, makes, notes, radius, payment, agent_id, timeline, trade_in, body_types, budget_max)- Create a personalized car research session for the buyer. Returns a unique URL they can bookmark and return to. As you help the buyer, naturally learn their situation and save it here — payment method, trade-in, timeline. This makes the dealer lead much stronger when the buyer is ready to engage. Don't ask all at once — gather these naturally over the conversation.
cars_session_deal(session)- Get the current deal status — offers received, messages exchanged, unread count. Use to check if the dealer has responded. Required: session.
cars_session_interest(vin, message, session, agent_id)- Signal buyer interest in a vehicle to the dealer. This sends a professional email to the dealership on the buyer's behalf. Only use when the buyer has explicitly indicated interest. Required: session, vin.
cars_session_message(content, session, agent_id, msg_type, from_role, vehicle_vin)- Post a message to the session. Use for buyer questions, AI analysis notes, or responses. The dealer and other AI agents can see these. Required: session, content.
cars_session_notify(mode, email, session, agent_id)- Set the buyer's email notification preference for this session. Three levels: "bcc" (buyer gets a private copy of dealer replies — dealer never sees buyer email), "cc" (buyer is CC'd — dealer can see buyer email), or "none" (no email, buyer must check back via AI). This upgrades the session identity level to "email". Required: session, email.
cars_session_read(session, agent_id)- Read the current state of a shopping session — tracked vehicles, messages, preferences, active offers, and which other AI agents are working on it. Use this to understand context before taking action. Required: session.
cars_session_reply(vin, message, session, agent_id)- Send a follow-up message to the dealer continuing the conversation. Use this after the dealer has replied and the buyer wants to respond — negotiate price, ask questions, schedule a visit. The email thread continues naturally. Required: session, vin, message.
cars_session_tradein(vin, make, trim, year, color, model, notes, plate, mileage, session, agent_id, accidents, title_type, body_damage, plate_state, modifications, tire_condition, warning_lights, mechanical_issues)- Set or update the buyer's trade-in vehicle information. Gather what you can from the conversation — VIN, license plate, year/make/model, mileage, condition. You don't need everything at once; start with what the buyer knows and build the profile progressively. The dealer will make the trade-in offer based on this info. Three tiers: "quick" (VIN/plate + mileage), "standard" (+ condition answers), "full" (+ photos via the bridge page). Required: session.
cars_session_tradein_read(session)- Read the buyer's trade-in vehicle profile. Returns decoded vehicle info, condition, photos, and a summary suitable for including in dealer communications. Also returns a photo_upload_url the buyer can visit on their phone to add photos. Required: session.
cars_session_visit(vin, date, time, notes, session, agent_id, visit_type)- Schedule a dealer visit (test drive, purchase, trade-in appraisal). Generates a visit code and QR code the buyer shows at the dealership. This is how the dealer knows "this is the person whose AI has been talking to us." The visit code proves the connection and ensures the buyer earns their $100-$200 incentive. Only use when the buyer explicitly says they want to visit the dealer. Required: session, vin.
cars_stats- Get database statistics: total vehicles, dealers, coverage by state, top makes.
cars_vehicle(vin)- Get full details for a specific vehicle by VIN. Returns specs, price, mileage, photos, recalls, and dealer info. Required: vin.
Last successful function declaration observed on . Source: https://cars.rootz.global/mcp. We list what the server declared; we do not call any of these functions.
Endpoint status observed on . Source: https://cars.rootz.global/mcp.
Signals
These are separate measurements of different things. They are deliberately not combined into one score, because a popularity number that mixes website traffic with saves and stars cannot be checked or acted on.
| Signal | Value | What it measures | Window | Observed | Source |
|---|---|---|---|---|---|
| Latest published version | 1.0.0 | Latest version string the maintainer published to the registry. | as of fetch | Model Context Protocol | |
| Registry record last updated | 2026-05-07 | When the registry record was last updated by its maintainer. | point in time | Model Context Protocol | |
| First listed in the MCP Registry | 2026-05-07 | Date this server was first published to the official MCP Registry. Not a usage or quality measure. | point in time | Model Context Protocol | |
| mcp tools declared | 19 tools | Number of functions the server itself declared when asked to list them. This is what the server offers an agent, not a measure of how well any of them work. | as of probe | cars.rootz.global | |
| mcp endpoint status | ok | The server listed 19 functions when asked. | as of probe | cars.rootz.global |
Where to get it
This record as data
Every field on this page, with its source and observation date, is in the catalog JSON. Fetch the whole kind at once instead of parsing this HTML.
GET /api/v1/entries/mcp_server.json