This card is populated live from a single CarVector request — image, specs, and federal recall count, with no scraping and no stitching five sources together. Ship a vehicle profile this afternoon.
You get year/make/model and a horsepower number — no image, no recall history, no diagnostic reference. CarVector bundles the data developers actually reach for, in one call and one key.
A growing catalog of vehicles by year, make, model, and trim — with representative illustrations, because a spec sheet with no picture isn’t much of a product.
Learn moreFederal recalls, owner complaints, technical service bulletins, and defect investigations — the full safety and failure history, mapped to vehicles and refreshed automatically. The history most vehicle APIs make you go find yourself.
Learn moreA growing library of diagnostic trouble codes with titles, categories, severity ratings, and safety flags. Look up any check-engine code by its identifier.
Learn moreSpecs and representative images, complete federal recall data, and a growing DTC reference — bundled in one API at a flat monthly, not per-seat. Start free.
Explore the API$ ▍
{ "id": "1laqdklflb3hfav", "year": 2018, "make": "Toyota", "model": "Tacoma Access Cab", "horsepower": 278, "displacement_l": 3.5, "transmission": "Automatic 6-spd", "drive_type": "4x2", "image_type": "illustration", "recall_count": 0 }
Tens of thousands of corrections and counting — each one a data point a single-source database gets wrong.
Drop CarVector into Claude, Cursor, or any MCP client and your agent queries real vehicle data — specs, recalls, and DTC reference — instead of hallucinating it. One config block, your key, four tools.
{
"mcpServers": {
"carvector": {
"command": "npx",
"args": [
"-y",
"carvector-mcp"
],
"env": {
"CARVECTOR_API_KEY": "cv_your_key"
}
}
}
}By the numbers
Last data refresh: 6 hours ago
We ship every week
See the full changelogLiveness probe. GET /health answers publicly on api.carvector.io with nothing but a status and a timestamp — no version, no build id, no dependency status, no row counts. It sits outside the paywall and outside key authentication on purpose, so a monitor polling it every minute can never book an authentication failure against your account. It is deliberately boring: a health check that is also reconnaissance is worse than none.
Pay-per-call is live on Base mainnet. You no longer need an account to buy data. An API call without a key now returns HTTP 402 with a signed price quote instead of a rejection — $0.01 in USDC on Base, quoted as 10,000 atomic units against the token's own EIP-712 domain so your wallet can verify what it is signing. Pay the quote, repeat the call with the payment attached, and the data comes back. Seven endpoints are priced this way: /v1/vehicles, /v1/vehicles/{id}, and that vehicle's /recalls, /tsbs, /complaints and /investigations, plus /v1/dtc/{code}. Quotes hold for 60 seconds. Every settled payment returns the payer their own receipt, and one payment buys exactly one call — replay protection is enforced at the ledger with a unique index, not by a time window, because a window alone was half a fix and we measured it.
The payment rail, built and deliberately switched off. Verification, settlement and serving all worked end to end before the flag was ever turned on, against testnet first and with the ledger reconciled after every round trip. Three things were wrong in the wire format and one of them was fatal, which is the argument for shipping it dark. We also taught the intrusion alarm what an attack looks like once payments exist: before this, a request with no API key was a probe, and afterwards it is the ordinary opening move of a paying customer. Counting those as attacks would have paged us about roughly a hundred customers per fifteen minutes. Key-guessing is still counted; arriving without a key is not.
The sweet spot for production apps.
Five tiers in all, including Enterprise. Full pricing
Building the next shop platform? Get specs, recalls, and DTC reference via API for a flat monthly instead of licensing legacy databases at hundreds per seat.
Learn morePull recall exposure and decode trouble codes at intake from the year/make/model. Structured data, not phone calls.
Learn moreMap open recall exposure across your fleet from each vehicle’s year/make/model. Know what’s outstanding before it’s a roadside breakdown.
Learn moreGive your agent structured automotive knowledge through MCP. No hallucinated specs or recalls — real data, real sources.
Learn moreOn the roadmap
Reset sequences, TPMS relearns, service light walkthroughs — derived from real shop content, not from manuals nobody publishes.
70 of 70 pilot vehicles cached
Paste a 17-character VIN, get back the decoded year, make, model, body class, and engine — joined to the matching catalog row's image and recall history. One call, complete picture.
Scaling from 3 procedure types (oil reset, TPMS, service reminder) to 10+ — covering key fob programming, throttle relearn, steering angle reset, and more. Coverage grows weekly.
Free tier. No credit card.
Not ready for a key? Browse the vehicle guide — every page is real data from the same API.