> ## Documentation Index > Fetch the complete documentation index at: https://zenorent.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Identifying a vehicle > Scan a VIN from a photo, trade a plate for one, or walk year and make — then price the glass and check who has it Before you can quote glass you have to know exactly which car it is. Zeno takes the three handles a customer actually gives you — a photo, a plate, or a description — and resolves all of them to the same vehicle record. This lives in **Glass → Parts**, on the **Lookup** tab. No job required. ## Scanning a VIN from a photo Customers send photos, not typed VINs. Zeno reads a VIN from any of them: * the VIN plate in the windshield corner * the door-jamb sticker * a registration card * an insurance ID card Photos straight off a phone work, including iPhone HEIC files. Keep them under about 12 MB. ### Why a scan gets rejected A VIN that *looks* right is the expensive failure in a glass shop: it reads as authoritative, nobody re-checks it, and the wrong glass ships. So a scanned VIN has to prove itself before Zeno will put it in the field. Three checks, widening: 1. **The check digit.** Free and instant. It catches almost every misread. 2. **A decoder round-trip.** This catches the character collisions the arithmetic structurally can't see — `J` against `1`, `S` against `2`. When the decoder can pinpoint the bad character, the rejection carries a suggested correction. 3. **Agreement with the document.** The year, make, and model printed on the page are read far more reliably than the VIN itself. If the VIN decodes to a 2019 Honda and the card says 2024 Jeep, something is wrong. Anything short of a clean pass comes back as a rejection with a reason, and you get the photo and an empty VIN box side by side rather than a filled-in guess. A blank field costs thirty seconds; a wrong VIN sends the wrong glass. **A bad read gets a second opinion.** When the first pass produces a VIN that fails its check digit, Zeno re-reads the photo with a stronger model — and accepts that answer only if it passes the check digit too. A second bad read never overwrites the first. **Illegal characters get repaired, not guessed.** A VIN can't contain I, O, or Q, so a read containing one is definitely wrong somewhere. Rather than assume — I is probably 1, O is probably 0 — Zeno works out every legal VIN the read could have been and keeps only the ones whose check digit passes. If exactly one survives, you get it as a suggested correction. If several do, you get none, because a plausible-looking wrong answer is the failure worth avoiding. That caution is earned. Substituting I→1 as a matter of course once turned a Mazda's genuine `J` into a `1` that passed every check — the two characters share a check-digit value, so nothing downstream could catch it. You'll see four outcomes: | Result | What it means | | - | - | | **Verified** | Decoded cleanly and agrees with the document. Use it. | | **Rejected** | Failed a check. The reason says which, sometimes with a suggested VIN. | | **Conflict** | Decoded cleanly but contradicts the year, make, or model on the page. Check you photographed the right car. | | **Unverified** | Check digit passed but the decoder was unreachable. Usable, unproven. | A decoder outage doesn't become a VIN outage — the read degrades to unverified on the strength of the check digit rather than blocking your work. ## When the photo shows a plate Half the photos a front desk gets never show a VIN — a rear-end damage shot, a claim photo, a plate close-up. The plate is a perfectly good handle. Enter the plate and its state. Zeno tries GlassBiller first, since that's already authenticated and costs nothing, and falls back to AutoBolt, which also returns year, make, and model with the VIN. The VIN that comes back is still only a claim, so it goes through the same three checks a scanned one does. A plate service can hand back a VIN that fails its own check digit, and decoding that lands the wrong glass just as surely as a misread photo. ## When you don't need a VIN at all Plenty of vehicles take the same glass across the whole model run. Walk **year → make → model → style** instead and skip the VIN entirely. It resolves to the same vehicle record the VIN would have. Reach for the VIN when the glass varies within a model — a windshield with a rain sensor, a heated wiper park, an ADAS camera bracket — because that's where the trim-level detail decides which part fits. If you already have the part in hand, the **know the part number?** search takes a bare NAGS number straight to the vendor grid and skips identification altogether. ## From vehicle to price and availability Once the vehicle is identified: 1. **Pick the shop.** GLAXIS accounts and warehouses are per-shop, so pricing and availability follow the location. 2. **Pick the glass position and part.** No need to re-walk the cascade the VIN already answered. 3. **Read the pricing and availability.** NAGS list and OEM pricing, plus a live three-vendor GLAXIS grid. 4. **Add to a vendor cart.** Carts are reviewed and ordered from the **Carts** tab. ## When to use AutoBolt instead The **AutoBolt** tab is a different question. NAGS identification answers "which vehicle is this, and what glass does it take." AutoBolt decodes a VIN or plate into **glass part numbers and the ADAS calibrations the windshield requires**. Go there when: * You need to know whether a windshield triggers an ADAS calibration before you quote it. The requirement is called out on the result. * NAGS identification is ambiguous and you want a second read on the part number. Two more things: * **A repeat decode is free.** Zeno stores every decode and serves the same VIN from its own record without calling AutoBolt again. * **The history is Zeno's.** AutoBolt has no history of its own, so every decode your shop has run is listed under the tab — searchable later. If AutoBolt returns several parts with different features, it says so rather than picking one. Confirm the trim against the car before you order. ## Where to go next * [Glass](https://zenorent.com/docs/glass.md) — the module, its rail, and how jobs tie together. * [Reports](https://zenorent.com/docs/reports.md#glass) — analytics and cross-shop job search. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.