Construction Contract Extractor

Extract the owner, contractor, project, contract sum, retainage, completion dates, and schedule of values from an AIA construction contract PDF.

The full guide: Schedule of values from an AIA A101 contract

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 Construction Contract Extractor.

What does the construction contract extractor read?

The contract number and date, the owner name and address, the general contractor name and address, the project name and site location, the architect name, the contract type (lump sum, cost-plus, unit price, or time and materials), and the total contract price with its currency.

Which schedule and payment terms come out?

The start and completion dates, the effective and expiration dates, the payment terms, the retention (retainage) percentage, the surety bond amount, and the insurance requirements and governing law are each returned as fields.

Does it track the contract sum against change orders?

Yes. The original contract sum, the net change by change orders, the adjusted contract sum to date, the total completed and stored to date, the retainage held, and the balance to finish come back, following the AIA G702 accounting.

Are the schedule of values and change orders returned as tables?

The schedule of values table returns each line item with its number, description, scheduled value, work completed previous and this period, materials stored, total completed, percent complete, and balance to finish, mapped to CSI MasterFormat divisions; a change order log table also comes back.

What are the file limits and privacy terms?

PDF only, up to 10MB and 100 pages. The contract is processed via the Talonic API for extraction, is not retained for training, and is not shared.

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.