Extracting data from operating authority certificates
FMCSA grants operating authority to for-hire carriers, brokers and freight forwarders on one federal template, worded differently for each type of authority. Its daily batches can run to several hundred. Freight brokers and commercial auto underwriters collect a copy from each carrier they onboard or insure, as a download or inside an emailed carrier packet.
A broker or underwriter needs the docket number, the USDOT number, the holder's legal name and the authority granted. Onboarding and underwriting systems check them against FMCSA's public registration records, confirming the authority is active and fits the operation being booked or insured.
What makes operating authority certificates hard
Many grants in one file
FMCSA publishes each day's grants as one PDF of near-identical one-page documents. Each page has to return as its own record.
Identifiers stacked by position
Docket number, holder name and city form a centered stack under the heading, told apart by their format and position on the page.
The grant sits mid-sentence
The authority granted is a bold phrase inside a sentence, such as common carrier of household goods, and its wording varies by authority type.
Parse, extract, verify
Parse
Parse turns the document into reading-order Markdown and a hierarchy of blocks (text, tables, figures), and grounds each block to its exact location on the page.
Extract
You define the fields in a JSON schema, and Extract returns their values in a consistent, structured format, each linked back to where it appears in the Markdown.
Verify
Each value links back to where it appears, down to the character range in the Markdown, and the page and bounding box in the source document, so every result can be verified.
One FMCSA operating authority certificate, end to end
About this document: Twenty operating authority documents issued by FMCSA's Office of Registration on a single service date, one per page. They come from the agency's published daily register. Most are common carrier certificates for general freight. Page 5 is a household goods certificate, page 8 a contract carrier permit, and three holders print a D/B/A line. The mailing envelope pages between documents were removed.
Processing objective: The objective is to turn each page into one carrier record, holding the identifiers a broker or underwriter checks against FMCSA's registration records. The featured page 1 is a common carrier certificate, the kind most of the batch shares, and its values read as one complete grant.
Demonstration fields: Service date, MC docket number, U.S. DOT number, Carrier name, Authority granted
The document
4 of 20 pages
Page 1shown below
Page 5
Page 8
Page 13
Where each value came from
authority_documents[0].service_date
{
"value": "February 16, 2023",
"page": 1,
"range": {
"end": 168,
"start": 150
},
"box": {
"xmin": 0.61743,
"ymin": 0.19185,
"xmax": 0.75182,
"ymax": 0.20628
}
}authority_documents[0].docket_number
{
"value": "MC-1179808-C",
"page": 1,
"range": {
"end": 194,
"start": 182
},
"box": {
"xmin": 0.40517,
"ymin": 0.26136,
"xmax": 0.52054,
"ymax": 0.27497
}
}authority_documents[0].usdot_number
{
"value": "3537401",
"page": 1,
"range": {
"end": 215,
"start": 195
},
"box": {
"xmin": 0.3805,
"ymin": 0.27951,
"xmax": 0.55144,
"ymax": 0.29308
}
}authority_documents[0].carrier_name
{
"value": "PINK SPARROW SCENIC LLC",
"page": 1,
"range": {
"end": 239,
"start": 216
},
"box": {
"xmin": 0.35135,
"ymin": 0.29402,
"xmax": 0.58282,
"ymax": 0.30803
}
}authority_documents[0].carrier_type
{
"value": "common carrier",
"page": 1,
"range": {
"end": 363,
"start": 253
},
"box": {
"xmin": 0.08526,
"ymin": 0.34837,
"xmax": 0.84502,
"ymax": 0.36514
}
}How to try it yourself
Everything on this page is reproducible. The sample, its schema, and the Parse and Extract API outputs are published. You can repeat the same two calls from whichever surface you already build on.
- Claude Skilldescribe the job; the agent writes it
- CLIno code at all
- MCPexpose ADE docs as an agent tool
- REST APIany language
- Pythonlandingai-ade
- TypeScriptlandingai-ade
Documents into data
Accurate
The best accuracy on the difficult documents other tools get wrong, such as messy scans, complicated tables, checkboxes and handwriting.
Auditable
Every answer cites its sources. Values come back with an audit trail (page, bounding box, character range) so a human can review them.
Agentic
Applies multi-step reasoning, and plans before it acts. Handles documents it has never seen before and adapts to variations.
API-first
Composable APIs that drop into any tech stack and can be called from anywhere. Easy to build on and embed in your pipelines and products.
Affordable
Cost-competitive rates with complete price transparency. Spend controls before you run, and every job reports its exact cost.
Enterprise-grade
Built for production and scale from day one. Fully observable and monitored, running securely in our cloud or yours.
Built for high-volume, high-throughput production
- Zero data retention
- Priority rate limits
- Enhanced support
- VPC and on-prem deployment
Available on Team and Enterprise plans. VPC and on-prem deployment require an Enterprise plan.
