Document type

Extracting data from wind mitigation inspections

Qualified inspectors record how a Florida home resists wind on the state's Uniform Mitigation Verification Inspection Form, with photographs attached as proof. Insurers and agents receive one for each policy seeking a windstorm discount, thousands a month across a book, as emailed PDFs or uploads to an agent portal.

Underwriting needs the answer to each of the nine questions, from building code and roof deck attachment to roof shape and opening protection. The rating engine maps those answers to the insurer's filed discount factors and applies the credit to the wind premium.

Start for Free Talk to Us

Page 1 of a sample Wind Mitigation Inspection, with extracted values outlined

25,376Markdown characters

201parse blocks

59fields extracted

The problem

What makes wind mitigation inspections hard

Answers are lettered checkmarks

Each question is answered by marking one lettered option, such as Roof Deck A through H, with sub-options nested beneath several of them.

A grid of marks

The opening protection chart crosses opening types, such as skylights and garage doors, against protection levels, and every answer is an X in a cell.

Options buried in fine print

Every option carries its code reference and construction detail in small type, so the marked answer sits inside a paragraph of qualifying text.

How it works

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.

Worked example

One wind mitigation inspection, end to end

About this document: A completed Uniform Mitigation Verification Inspection Form for a one-story Florida home, published by a licensed inspection company as a sample with fictional client details. Six portrait pages with a text layer carry the nine questions, the opening protection chart and the inspector's certification. Page 1 is featured. The cover and photo pages of the original file were removed.

Processing objective: The objective is to capture what an insurer reads first when pricing a wind discount. The inspection date matters because the form stays valid for up to five years. The building code and wind region answers are checked options, each grounded to its line on page 1.

Demonstration fields: Inspection date, Year of home, Building code answer, Wind region, Roof permit application date

The complete example is on GitHubThe sample document, the full extraction schema, the complete parse output and every extracted field.

Open the folder

The document

4 of 6 pages

Page 1shown below
Page 2
Page 3
Page 4

Where each value came from

Page 1 of the sample Wind Mitigation Inspection, with every featured field outlined

Inspection dateinspection_date

2026-07-29

The region of the sample Wind Mitigation Inspection containing Inspection date, with the extracted value outlined

Printed on the page as Inspection Date: 07/29/2026, normalized on extraction.

extraction_metadata["inspection_date"]

{
  "value": "2026-07-29",
  "page": 1,
  "range": {
    "end": 196,
    "start": 169
  },
  "box": {
    "xmin": 0.06602,
    "ymin": 0.06605,
    "xmax": 0.93363,
    "ymax": 0.09328
  }
}

Year of homeyear_of_home

1,999

The region of the sample Wind Mitigation Inspection containing Year of home, with the extracted value outlined

extraction_metadata["year_of_home"]

{
  "value": 1999,
  "page": 1,
  "range": {
    "end": 654,
    "start": 636
  },
  "box": {
    "xmin": 0.06613,
    "ymin": 0.26047,
    "xmax": 0.32064,
    "ymax": 0.28902
  }
}

Building code answerbuilding_code

The region of the sample Wind Mitigation Inspection containing Building code answer, with the extracted value outlined

extraction_metadata["building_code"]

{
  "value": "D",
  "page": 1,
  "range": {
    "end": 2449,
    "start": 2371
  },
  "box": {
    "xmin": 0.05574,
    "ymin": 0.52268,
    "xmax": 0.58755,
    "ymax": 0.53737
  }
}

Wind regionwind_region

Region 1

The region of the sample Wind Mitigation Inspection containing Wind region, with the extracted value outlined

extraction_metadata["wind_region"]

{
  "value": "Region 1",
  "page": 1,
  "range": {
    "end": 2611,
    "start": 2585
  },
  "box": {
    "xmin": 0.05591,
    "ymin": 0.56518,
    "xmax": 0.68309,
    "ymax": 0.58193
  }
}

Roof permit application dateroof_coverings[0].permit_application_date

2019-04-04

The region of the sample Wind Mitigation Inspection containing Roof permit application date, with the extracted value outlined

Printed on the page as 04/04/2019, normalized on extraction.

extraction_metadata["roof_coverings[0].permit_application_date"]

{
  "value": "2019-04-04",
  "page": 1,
  "range": {
    "end": 3509,
    "start": 3499
  },
  "box": {
    "xmin": 0.24217,
    "ymin": 0.73096,
    "xmax": 0.37642,
    "ymax": 0.74927
  }
}

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

Run this sample yourself This document on GitHub

Why LandingAI?

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.

Compare plans

Capabilities

Wind Mitigation Inspection processing highlights

Nine lettered question answers

Nested sub-answers

Building code and wind region

Roof deck attachment

Roof to wall connection

Roof geometry

Secondary water resistance

Opening protection chart

Permit application dates

Inspector certification block

No per-inspector templates

Bounding box for every value

FAQs

Questions, answered

Yes. Parsing writes every checkbox into the Markdown in one standard form, marked checked or unchecked beside its label. A field for each question comes back as the letter of the option marked, with any sub-answer beneath it.

Yes. Describe each opening type as a field in your schema. Windows, garage doors, skylights and the other opening types each come back with the protection level marked in their column of the chart.

Typed entries such as the inspection date and permit application dates come back as values in your schema. Dates return in one consistent ISO format, while grounding still points at the string as the inspector typed it.

No. A schema names the fields you want and describes them in plain language. Extraction reads the parsed form for meaning, so one schema covers every inspector's completed version of the state form.

Every extracted value carries the character range it was quoted from in the parsed Markdown. The Ground API maps that range to a page and a bounding box, so a reviewer sees each answer on the line that printed it.

Start extracting wind mitigation inspection data

Reliable, structured outputs with full traceability in minutes