Closing Disclosure Extractor

Extract the borrower, seller, loan terms, purchase price, closing costs, cash to close, and prorations from a Closing Disclosure or settlement statement.

The full guide: Cash to close and loan costs on a Closing Disclosure

How should we start?

Build with Talonic

Need to scale? Create an API key, then run this from your own code or an agent.

Create an API key

Free account required

Start with a document

Upload a file or pick a sample, and see the fields come back.

No signup · nothing stored

Questions about Closing Disclosure Extractor.

What does the closing disclosure extractor read?

The closing disclosure number and date, the note date, the subject property address and type, the borrower name, the seller name, the lender name, the loan identifier, and the title company handling the closing.

Which loan and price figures come out?

The loan amount, interest rate, loan term, and maturity date, alongside the purchase price, the appraised value, the down payment, and the loan estimate reference number, so the disclosed loan can be compared against the estimate.

Does it total the closing costs and cash to close?

Yes. The total closing costs, the costs paid by the buyer, the costs paid by the seller, the net prorations and adjustments, and the cash to close required from the buyer all come back as fields.

Are the cost line items returned as tables?

A closing costs table returns each cost type, description, amount, and paying party; a loan costs table returns origination and discount-point items; a prorations and adjustments table returns each item, its period, daily amount, total, and the party credited.

Does the tool give mortgage or financial advice?

No. It only extracts the figures printed on the disclosure into structured data and offers no financial, mortgage, or legal advice. PDF only, up to 10MB and 100 pages; the document is processed for extraction only and is not retained.

Doing this to one file, or to ten thousand?

The tool reads a single document. The platform reads the whole estate once and keeps it queryable — the same engine, with a memory.

See PDF to Markdown if you run this for data and platform teams, or the extraction API if you are building it in.