ingest_agreement
Hand Aurgus your rebate agreement — PASTE its text, or DROP a PDF/scanned document (pass it as file_base64). Aurgus reads the terms with its deterministic parser (and CV/OCR for a PDF/scan — a Python CPU service, ZERO LLM): customer, basis (whole-volume / incremental / cliff), rate or tier ladder, and the rebate period. A CPU parser proposes spans, deterministic rules produce every value, and anything it can't read cleanly is flagged for your eye rather than guessed. Nothing is created until a person reviews and certifies the terms in the browser — this is how the agreement (and the ASC 606 treatment it implies) gets set up, before any transaction data is asked for.
Tool call template
This tool declares no required parameters. Add optional parameters from the input schema as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {},
"name": "ingest_agreement"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| text | No | The agreement's text (paste the full agreement, or the rebate clauses). Provide this OR file_base64. | |
| filename | No | The original filename (e.g. 'rebate-agreement.pdf') — used to sniff the file type and shown on the review page. | |
| file_base64 | No | A PDF or scanned-image agreement, base64-encoded. Aurgus reads it with CV/OCR (zero LLM). Provide this OR text. |