Wenmar Open

Free VIN decoder and vehicle catalog from NHTSA vPIC data.

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

    What data it sees

    Do you need an account

    No: the server works without sign-in

    Free VIN decoder and vehicle catalog from NHTSA vPIC data. Read-only, no key.

    How to connect

    How to connect Wenmar Open to Claude

    The server address to paste into Claude settings:

    https://open.wenmarpro.com/mcp

    1. Open Claude (claude.ai or the desktop app).
    2. Go to Settings → Connectors.
    3. Click Add custom connector.
    4. Paste the server address copied below into Remote MCP server URL and click Add.
    5. In a chat, click + → Connectors and switch the new connector on.

    Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

    Authorization

    After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

    How to connect Wenmar Open to ChatGPT

    The server address to paste into ChatGPT settings:

    https://open.wenmarpro.com/mcp

    1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
    2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
    3. Open Settings → Connectors and click Create.
    4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
    5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
    6. In a new chat open + → Apps/Connectors and enable the connector.

    OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

    Authorization

    On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

    How to connect Wenmar Open to Cursor

    The server address to paste into Cursor settings:

    https://open.wenmarpro.com/mcp

    Fastest: click Open in Cursor below and confirm the prompt.

    Manually:

    1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
    2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
    3. Save the file. The server appears in the MCP list; authorize it there if asked.

    Authorization

    If the service needs sign-in, an authorize button appears next to the server in the MCP list.

    Server tool list (6)

    Raw names from tools/list. Only developers need these.

    wenmar_vinDecode vehicle identification numbers into year, make, model, trim, engine, drivetrain, safety equipment and the matching catalog entry. action=decode takes `vin` (and optionally `year` to override the model year). action=batch takes `vins`, at most 50, and returns one result per VIN in order. A wrong check digit is not an error: the result has valid=false and a warning. action=overview takes `vin` (and optionally `year`) and gives everything held for the model year the VIN decodes to in one answer: the decode in brief, a count of the recall campaigns that may apply to that year, make and model with the three newest, and links. A block with no data is null. action=fuel, specs, fluids or tires takes `vin` (and optionally `year`) and gives that data for the model year the VIN decodes to: fuel consumption ratings from EPA and Natural Resources Canada (optional `country`, us or ca), specifications from EPA's files, fluids (the engine oil grade EPA records, and records a person read from the maker's own documents, each with its source and the day it was checked), or the original-equipment tire fitments. When the decode settles the engine, fuel, specs and fluids hold only what is for that engine and `engine_settled` is true; tires are never narrowed.
    wenmar_vehiclesLook up vehicles by year, make, model, submodel and engine. Actions: years; makes (optional `year`); models (`make`, optional `year`); submodels (`make`, `model`, `year`); engines (`make`, `model`, `year`, optional `submodel`); search (`query`, such as "2019 civic si" or "chevy 1500"); entry (`id`, such as "2019_honda_civic_si"); overview (`id`: everything held for that model year in one answer, with a count of the recall campaigns that may apply and the three newest; a block with no data is null); model_years (`make`, `model`: every model year of one model, newest first, each with its vehicle id and the number of recall campaigns that may apply); safety (`id`: NHTSA's crash test ratings for that model year, one entry for each vehicle NHTSA rated, by body style and drive, each with its stars; an entry with no stars is a vehicle NHTSA lists and has not tested); fuel (`id`, optional `country` us or ca: fuel consumption ratings from EPA and Natural Resources Canada, each saying in `confidence` how far it was matched: model, engine or submodel); specs (`id`: transmission, drive, recommended fuel and more from EPA's files, one entry per EPA rating); fluids (`id`: the engine oil grade EPA records for each engine, and records a person read from the maker's own documents, each with its `source` and the day it was `checked`; an empty list means nobody has recorded it); tires (`id`: every original-equipment tire size on record with the trims it covers; none is picked). `make` and `model` take a name, an alias or an id. `term` narrows a list to names starting with it. `scope` is light (cars, MPVs and trucks; the default), all, or a vPIC vehicle type id. action=complaints takes `id` and gives what owners reported to NHTSA about that model year as counts by component, with the median mileage owners gave: these are owners' reports, not verified, and a count is not a rate.
    wenmar_recallsRecall campaigns from NHTSA (United States) and Transport Canada that may apply to a vehicle. A campaign lists a year, a make and a model, so every answer has may_apply: true. It may apply; confirm by VIN with the manufacturer, or at the addresses in `lookup`. This is never a lookup by VIN. action=vehicle takes `id`, a vehicle id such as "2018_ford_f-150". action=vin takes `vin` (and optionally `year`): it decodes the VIN and lists the recalls of its model year. action=campaign takes `source` (nhtsa or tc) and `campaign`, such as "19V888000", and gives that campaign with every vehicle it lists. action=search lists campaigns newest first, narrowed by any of `make`, `model` (needs `make`), `year`, `component` and `query` (free text); `limit` is at most 100 and 25 when left out; `cursor` is `next_cursor` of the page before. action=recent takes `since`, a day such as "2026-09-01", and lists the campaigns reported on or after it, newest first, each with the vehicles it lists; it also takes `make`, `limit` and `cursor`. `country` is us or ca on every action but campaign; left out, both.
    wenmar_bulletinsManufacturer communications NHTSA publishes: service bulletins, service campaigns, warranty extensions and over-the-air updates. Each has NHTSA's summary, the manufacturer's own document number, and `document`, the address NHTSA's copy of the document is expected at (not confirmed; null for old ones). The document itself is the manufacturer's and is not returned. A bulletin lists model years: it is not a recall and does not say a vehicle is affected. action=vehicle takes `id`, a vehicle id such as "2018_ford_f-150", and gives one page of that model year's bulletins, newest first, with `counts` by type, component, trouble code and topic. action=vin takes `vin` (and optionally `year`) and does the same for the model year the VIN decodes to. action=get takes `nhtsa_id` and gives one bulletin with every vehicle it lists. action=search lists all bulletins newest first, narrowed by `make`, `model` (needs `make`) and `year`. Every action but get takes `type` (service_bulletin, service_campaign, warranty_extension, over_the_air, emissions, other), `component` (an id from `counts.components`, such as brakes), `code` (a trouble code such as P0420), `topic` (maintenance), `query` (free text in the summary, such as a symptom), `limit` (at most 100, 25 when left out) and `cursor` (`next_cursor` of the page before). This tool needs the hosted service: a local data file does not hold bulletins.
    wenmar_codesGeneric OBD-II trouble codes, described in Wenmar Open's own words: a title, what the code means, common causes, first checks, the components it concerns and related codes. Only codes every maker uses the same way are described (P0, P2, P3400 and up, B0, C0, U0, U3). Manufacturer-specific codes are not described. action=get takes `code`, such as "P0420", and gives its description. action=search lists the described codes by code, narrowed by `query` (text in the code, title or meaning) and `system` (powertrain, body, chassis, network, or an area such as powertrain/emissions); pass `next_cursor` back as `cursor`. action=vehicle takes `id`, a vehicle id such as "2018_ford_f-150", and `code`, any code, a manufacturer-specific one included: it gives the description (null when there is none), the bulletins of that model year whose NHTSA summary names the code, and the bulletins and recalls of that model year about the same components. A part of that answer is null where the data is not held. A description is general and is not a diagnosis; a bulletin or recall listed may apply to the model year and is not confirmed for one vehicle.
    wenmar_laborWenmar's own estimates of the hours a repair job takes: written by people who do the work and later measured from repair orders, never from a labour guide. An estimate, not a quote. Rust, seized fasteners and added equipment change the time. action=jobs lists the jobs an estimate can be about; with `job` it gives one job: what it includes, what it excludes, and its typical hours by class of vehicle. action=vehicle takes `id`, a vehicle id such as "2018_ford_f-150" (an id that ends with an engine asks for that engine), and optionally `job`: for each job it gives `hours` (low, typical, high), `basis` (expert or observed) and `scope`, which says what the figure is for: engine, generation, model or class. A class figure is typical for vehicles of that class and was not worked out on this vehicle: say so. A job with no figure for the vehicle is not listed. action=vin takes `vin` (and optionally `year`, `job`) and does the same for the vehicle the VIN decodes to. No answer holds a price: multiply by the shop's own rate. Credit the data as `attribution` says: Labour estimates by Wenmar Pro, CC BY 4.0.