Document Types
Define what to pull out of your documents — the pipeline reads each file, extracts your fields, checks them, and saves the results to Excel, the database, or your email.
Loading…
Define what to pull out of your documents — the pipeline reads each file, extracts your fields, checks them, and saves the results to Excel, the database, or your email.
Loading…
No records yet.
Just type a Label like "Total Amount" — the field name (total_amount) is created automatically. Must exist = document is rejected if this field is missing. At least / At most = optional limits for numbers (e.g. At least 0.01 rejects a zero amount; At most 10000 catches misreads). Leave empty for no limits.
Documents come from My Files. A folder elsewhere on this computer appears here only after you open it under Settings → Security → Folders outside AIK.
Saved with the type — the Run button uses this automatically. Upload documents on the My Files page, or choose "Custom folder path" to point at any folder on the server / your machine.
Uses the mailbox in Settings → Mail. Only e-mails that arrive after you switch this on; each one is handled once. Nothing is marked as read. Attachments are saved in My Files → inbox/<date>.
Every valid record is sent as JSON. Leave URL empty to disable.
Write my database for your own database (database.db in My Files — AIK Chat and the SQL agent read it), or a PostgreSQL / MySQL address. The table is created if it does not exist.
While on trial the agent reads the documents but saves, sends and moves nothing. It compares every answer with the spreadsheet the person already keeps, and Go live stays locked until it has been right enough times in a row.
Spreadsheet · match on · correct in a row needed to go live
Pick a folder of sample documents and the spreadsheet with their correct values. Each model reads every document — nothing is saved, sent or moved — and you see how many fields it got right, what it cost and how long it took. Up to 3 models and 20 documents.