fileScout turns a file into structured data. It finds each element, such as text, tables, charts and images, reads each one the way that suits it, and fills the fields of your AI Schema. Financial reports, claim forms and contracts come back in the same shape every time.

annual-report-2025.pdf$0.015 per page

→ Every field of your AI Schema, as JSON

​
Start here

Quickstart

Upload a file and read its values in five minutes.

Try the Sandbox

See the fields on the Developer Portal, with no code.

API reference

Every upload, file and schema endpoint.

​
How it works

1

Map the elements

fileScout records each element and where it sits: headings, paragraphs, tables, lists, form fields, charts, images, logos, signatures, handwriting, stamps and QR codes.
2

Read each element its own way

Tables keep their rows and columns. Text is read with its layout. Charts, photos and stamps are read visually, and scanned pages go through OCR first. One file can use several models and tools.
3

Skip what carries no data

Running headers and footers are left out, so they cost no processing.
4

Fill your schema

The results come together in the fields of your AI Schema, returned by GET /v1/files/{fileId}/values.
Each element gets the method that reads it best, so accuracy stays high on mixed files and token use stays low. The element breakdown is internal; the API returns the structured values. See it on the 60-page sample above in Multi-page financial report, or download the sample and run it yourself.

​
What’s included

​
Turn it on

Add FILE_SCOUT to enabledProducts when you request an upload:
fileScout can’t be combined with storageOnly: true. To check the same file with fileForensics, add FILE_FORENSICS too; see Choose products.

​
Pricing

fileScout costs $0.015 per page, counted once processing completes. See Billing.

​
Next steps

Extract fields

Define the fields fileScout fills.

Financial report example

A 60-page report, element by element.