Extract from a document set

Define the fields you need once, attach a document, and Anara returns a structured output using only what that document says. Run the same prompt against every document in the stack, one chat per document. Use it when you process the same document type repeatedly and need a consistent, auditable extraction.

Prompt

A regulatory compliance analyst works through a backlog of Brazilian administrative infraction tickets.

Extract the following fields from the attached document. Use only what the document explicitly states. If a field is not present in the document, write "Not found" for that field. Do not infer or estimate any value.

Fields to extract:
- Case number
- Date of infraction
- Entity cited
- Violation type (quote the specific article or clause cited)
- Penalty amount (exact figure as stated)
- Deadline for response or appeal

After the extraction, quote the exact passage from the document where the penalty amount appears.

What to add to your library

• Required: the document to extract from, uploaded to your library or attached to the chat.
• Optional: a completed example extraction from a prior document of the same type, so Anara matches your format and detail.
• Optional: the document type and jurisdiction, if the fields depend on legal or regulatory conventions not obvious from the text.

What you get

• Each field filled with the value the document explicitly states, or "Not found."
• A source quote for your most important field, as an audit trail.
• An output you can paste into your tracking spreadsheet before opening a new chat for the next document.

Follow-ups

Confirm the document type before extracting.

Before extracting any fields, identify the document type and confirm whether this document is a formal infraction notice, an administrative warning, or a pre-infraction notification. Then proceed with the extraction only if it is a formal infraction notice. If it is not, stop and tell me what it is instead.

Compare two documents for the same case.

I have two documents for the same case. Extract the penalty amounts and response deadlines from both and compare them. Flag any discrepancy between what the two documents state.

Tips

• Treat the prompt as a versioned protocol: test it on three or four documents, fix wrong or missing cells, and keep a record of each change.
• Once you know your edge cases, add a "CRITICAL INITIAL VERIFICATION" step at the top that checks for them before extracting.
• If the extracted value does not appear in the quoted passage, the extraction failed. Run it again.