What data does VehDB provide?

VehDB covers cars, motorcycles, power boats and tire fitments with VIN-grade specifications, enriched with U.S. public data: EPA fuel economy (MPG, EV range, fuel cost, CO2) and motorcycle engine certification, NHTSA safety data (recall campaigns, NCAP crash-test ratings, owner complaints, service bulletins and warranty extensions, tire recalls), U.S. Coast Guard boat recalls and builder codes, and EIA fuel and electricity prices. Tire fitments include computed geometry and alternate-size validation against the industry ±3% diameter rule.

How can my business access the data?

Three ways, same dataset: a search workspace in the dashboard for your team, a REST API for your systems, and an MCP server so AI agents like Claude can query vehicle data as native tools. The free plan includes API access with 100 calls per month — no credit card required.

Is there an API for vehicle recalls and tire fitment?

Yes. GET /v1/cars/{uuid}/recalls returns NHTSA recall campaigns and crash-test ratings; GET /v1/tire-sizes/{uuid}/geometry returns computed tire geometry with alternate-size diameter deltas; GET /v1/cars/{uuid}/tire-sizes cross-references which tires fit a car.

VehDB is operated by Savas OÜ, an EU company registered in Estonia. Billing runs on Stripe and personal data is handled under the GDPR.

Enriched with EPA, NHTSA, USCG & EIA public data

Vehicle data your business
can build on.

Cars, motorcycles, boats and tire fitments — VIN-grade specs, fuel economy, recalls and fitment math. One dataset, three interfaces: search workspace, REST API and an MCP server for AI agents.

Start free — no card required API docs

Free plan includes 100 API calls/mo + the full search workspace.

New: VehDB is in Claude's connector directory. Add it in one click

api.vehdb.com

# Specs + EPA economy in one call

GET /v1/cars/{uuid}

{ "make": "Hyundai", "model": "Tucson", "year": 2026,

"mpg_combined": 28, "annual_fuel_cost": 2200, … }

# NHTSA recalls, live

GET /v1/cars/{uuid}/recalls → "recall_count": 4

# Will that tire fit? Computed Ø delta

GET /v1/tire-sizes/{uuid}/geometry

{ "size": "245/40R17", "diameter_delta_pct": +4.66 } ✗ out of ±3%

# Same data for AI agents

MCP mcp.vehdb.com/mcp → search_vehicles · my_lists · tire_sizes_for_vehicle

Built for parts & fitment sellers, insurers, marine services and fleets across the US market.

1,342,720 vehicle records 3 interfaces — UI · API · MCP EPA · NHTSA · USCG · EIA enrichment

Now in Claude's connector directory

Ask Claude about any vehicle.

Add VehDB to Claude and it answers from our data instead of memory: recalls under every spelling NHTSA uses, the best and worst model years, EPA fuel cost, tire fitment and crash records. Works in claude.ai and the Claude apps.

Add to Claude How it works

1 Click Connect
2 Sign in to VehDB, approve
3 Ask. Free plan: 100 calls/mo

Claude VehDB connected

Any recalls on a 2021 Tesla Model S?

VehDB · recalls_for_vehicle

23 recall campaigns, filed by NHTSA under "MODEL S" (VehDB matches every spelling). The newest, 26V283000, covers back over prevention: software. 5 NHTSA investigations are still open.

Computed from today's VehDB data.

Made for the way your team actually works

Every feature below ships today — on the free plan or one click up.

Parts & fitment sellers

Answer "will it fit?" with math, not guesswork.

Tire geometry computed from any size — Ø, sidewall, revs/mile
Alternate sizes validated against the ±3% diameter rule
Reverse search: a size → every vehicle it fits
Export your shortlist to CSV for your price lists

Insurance & claims

Risk-relevant data on the vehicle, not just the spec sheet.

NHTSA recall campaigns per vehicle — live, via API
NCAP crash-test ratings (overall, front, side, rollover)
VIN-grade body, engine and drivetrain identification
Bulk-friendly REST API with plan-based rate limits

Fleets & operations

A working list your whole stack can read.

Lists: curate vehicles with notes (part numbers, refs)
Recall tracking: every NHTSA campaign hitting your lists — one dashboard, plus a daily email when a new one lands
Read the same list via GET /v1/lists or the my_lists MCP tool
Saved searches with email alerts when new records land
EPA fuel costs for TCO comparisons

Developers & AI builders

Ship vehicle features without building a dataset first.

REST API: filter, paginate, cross-reference
MCP server — Claude, Cursor or any agent queries it natively
OAuth2 for AI providers + simple Bearer tokens
Interactive playground with an auto-provisioned token

Fleet Management + AI

Run your fleet from a list.
Let AI read it.

Save the vehicles your business runs on — statuses, notes, done. We watch NHTSA for recalls on every listed vehicle and email you in the morning digest. Your systems read the same list over REST; your AI agent reads it over MCP.

daily recall digest GET /v1/lists/recalls my_lists · MCP

Fleet management → Fleet management with AI →

Your AI client · VehDB MCP connected

Which of our owned vehicles have open recalls?

2 of your 4 owned vehicles have open NHTSA campaigns — the 2021 F-150 has two, latest posted this week. The morning digest email already went to the fleet owner.

via my_lists + GET /v1/lists/recalls

One dataset. Three interfaces.

Your analyst searches it, your backend queries it, your AI agent calls it — same records, same IDs.

1 · Workspace

Search like a pro tool, not a brochure

Type “1999 Yamaha Exciter 220” and it just works — year-aware, cross-type search with a year filter right in the bar, a filter rail per catalog, dense result tables, shortlist CSV export, saved searches with alerts. Built for people who look things up fifty times a day.

Open the workspace →

2 · REST API

Everything is an endpoint

curl "https://api.vehdb.com/v1/cars?make=Toyota" \
  -H "Authorization: Bearer TOKEN"

Specs, lists, recalls, tire geometry, cross-references. Free tier included.

Read the docs →

3 · MCP for AI

Your agent already knows how to use it

{ "mcpServers": { "vehdb": {
  "url": "https://mcp.vehdb.com/mcp",
  "headers": { "Authorization": "Bearer TOKEN" } } } }

search_vehicles · get_vehicle · recalls_for_vehicle · complaints_for_vehicle · service_bulletins_for_vehicle · tire_sizes_for_vehicle · tire_dot_check · decode_hin · energy_prices · my_lists. OAuth2 for AI providers.

Connect an agent →

Not just specs.
Decision-grade data.

We enrich the catalog with U.S. government public data and computed engineering values — so the answer your team needs is already in the record.

EPA fuel economy: MPG, EV range, annual fuel cost and CO₂ — sourced from fueleconomy.gov.
EIA energy prices: Weekly gasoline and diesel pump prices and monthly electricity prices by state, so fuel cost runs at this week's price.
NHTSA recalls & NCAP: NHTSA recall campaigns and crash-test ratings per vehicle.
Service bulletins & warranty extensions: NHTSA manufacturer communications per car and motorcycle model year: service bulletins, special coverage and over-the-air updates.
USCG boat recalls & HIN decoding: U.S. Coast Guard recall campaigns by builder code, and any hull identification number decoded to its builder and build date, with the accidents reported under the hull.
Computed tire geometry: Diameter, sidewall, revs/mile and alternate-size validation — pure math, every size.
Tire DOT & recall checks: A DOT code read to its plant and build week, checked against the production windows in NHTSA tire recalls.

A single record, enriched

2026 Hyundai Tucson	SUV · AWD · GAS
MPG combined	28	EPA
Annual fuel cost	$2,200	EPA
CO₂ tailpipe	322 g/mi	EPA
Recall campaigns	4 active	NHTSA
OEM tires	235/65R17 — Ø 29.0"	COMPUTED

U.S. federal data is public domain — clean licensing, no lock-in.

Curate once. Read everywhere.

Your team builds a vehicle list in the dashboard — with notes, part numbers, statuses. Your pricing service reads it as JSON. Your AI agent reads it as a tool. No sync jobs, no exports. And when NHTSA posts a recall for a listed vehicle, it lands in the next morning's digest.

List worklist

"2024 Camry — PN-4421 fits, quoted for Jensen A/S"

GET /v1/lists → your systems

my_lists → your AI agent, via MCP

recall digest → your inbox, every morning

AI-NATIVE · MCP SERVER

Your AI agent's vehicle-data tools —
not glue code.

VehDB ships a native Model Context Protocol server. Claude, Cursor or any MCP client queries cars, recalls, fuel economy and tire fitment as first-class tools — no wrappers, no scraping, no glue code to maintain.

Native MCP tools — search_vehicles, recalls_for_vehicle, complaints_for_vehicle, tire_sizes_for_vehicle, my_lists — callable by Claude, Cursor or any agent.
One-click AI connector in Claude; in any other MCP client, paste one URL and sign in: no token to copy. Scripts use an API token. Same data and quota as the REST API.
No scraping, no function-calling boilerplate, no schema drift — the agent gets typed, documented tools.

Explore the MCP server Connect your agent

One-click AI connectors

Add to Claude

claude_desktop_config.json

{ "mcpServers": {
    "vehdb": { "url": "https://mcp.vehdb.com/mcp" }
  } }

# the agent calls a typed tool, not your scraper

tool search_vehicles(make="Toyota", year=2026)

tool tire_sizes_for_vehicle(uuid) → "235/65R17" …

AI-assisted research

Ask about any vehicle. Get an answer backed by records.

Connect VehDB to Claude, ChatGPT or Cursor and ask in plain words: recalls, the best model year, fuel cost, tires, crash records. These answers come from today's data.

Any recalls on a 2021 Tesla Model S?

Calls recalls_for_vehicle

23 recall campaigns, filed by NHTSA under "MODEL S" (VehDB matches every spelling). The newest, 26V283000, covers back over prevention: software. 5 NHTSA investigations are still open.

Recall campaigns: 23
Open investigations: 5
Newest: 2026-05-04