Skip to main content
TruckCodex
Hubs

About TruckCodex

Public trucking data joined for booking and vetting — free to browse, with transparent ads or partner offers when shown.

Free to browse · transparent monetization

TruckCodex is free to use without a paywall or TMS upsell. We may show advertising, affiliate links, or partner lead offers on some pages — always with disclosures (see Partners and our Privacy Policy). The mission stays the same: owner-operators, small brokers, and sales teams get the same joined FMCSA / NHTSA read that paywalled tools sell.

What TruckCodex is

A decision surface for U.S. trucking public data. Look up a USDOT, VIN, inspection, authority docket, or insurer exposure and get one honest read — status, insurance, safety signals, and history. We cross-join FMCSA, NHTSA, EIA, and related registries so the answer is citable and machine-readable.

Derived products go beyond a raw mirror: booking verdict heuristics, insurance-lapse watch, chameleon-carrier watch, email/phone fraud check, freshness and page-inventory transparency, plus original Knowledge articles.

Who publishes this

Publisher: Truck Codex (TruckCodex) — independently operated reference platform for U.S. trucking public data. Product priorities come from real booking and carrier-vetting work.

Postal address: 619 S 1st St, Patterson, CA 95363, United States

Accountability: editorial principles below, monitored inboxes, and the pages at /contact/, /corrections/, and /meta/sources/.

Why we exist

FMCSA, NHTSA, EIA, and IFTA publish enormous amounts of trucking-industry data, but it lives in siloed registries. The useful queries — "is this carrier bookable today?", "has insurance lapsed?", "what changed since last month?" — require joins and history. TruckCodex fills that gap with a structured, never-deleting surface a human can verify and an AI overview can attribute.

Editorial principles

  • No opinion as fact. Heuristics (chameleon score, insurance-watch tier, booking verdict) are labeled — not legal determinations.
  • No scraped secondhand data as primary. Fields trace to official upstream registries at /meta/sources/.
  • No synthetic numbers. Null upstream renders as "not published."
  • Every claim sourceable. Entity pages surface provenance.
  • Never delete history. Prior insurers, OOS, officers, and authority changes are retained.
  • Decision tool, not page farm. We noindex thin leaves when they add noise.
  • Honest monetization. Ads and partners are disclosed; we do not disguise lead walls as search results.

Methodology & corrections

Methodology · Sources · Freshness · Corrections · Partners & advertising

Contact

General: [email protected]
Corrections: [email protected]
Privacy: [email protected]
Full contact page: /contact/

Last reviewed: .