# TradepilotUSA Invoice Text Extraction

> TradepilotUSA Invoice Text Extraction is a paid API for AI agents from www.tradepilotusa.com, paid per call via x402, $0.25/call, status unknown (last checked 2026-10-02).

Extracts structured key fields (invoice number, seller, buyer, dates, line items, totals) from raw invoice text and returns them in a structured table with evidence gaps and review flags.

## Facts

- Endpoint: POST https://www.tradepilotusa.com/api/agent-commerce/v1/services/invoice_text_extraction/execute?utm_source=zero.xyz
- Price: $0.25/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/tradepilotusa-invoice-text-extraction-d895ea7b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_Zi3xK02WRqhfI1ZsEYWtd

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability tradepilotusa-invoice-text-extraction-d895ea7b -d '<json body>'
```

Example prompt: I have some raw invoice text and I need you to extract all the key fields from it — things like the invoice number, seller, buyer, dates, line items, and totals. Here's the text: 'Invoice #INV-2045 from Acme Hardware Co. to Summit Doors LLC, dated 2026-10-15. 3x steel hinges at $12.00 each, subtotal $36.00.'

## When to prefer this

Choose this endpoint when you have raw, unstructured invoice text (copied from PDFs, emails, or documents) and need to extract structured fields like invoice number, seller, buyer, dates, line items, and totals. It is especially useful when you also want explicit flags for missing or ambiguous fields and audit-ready provenance metadata. Prefer this over generic OCR or document parsers when the text is already in string form and you need invoice-domain-specific field identification with gap analysis.

## Known failure modes

- Source text too short (below 20 characters) returns validation error
- Source text too long (above 32,000 characters) returns payload too large error
- Highly ambiguous or corrupted invoice text may result in many evidence_gaps and requires_human_review=true
- Missing required source_text field returns schema validation error
- Non-invoice text submitted may produce empty or nonsensical field extractions

## How this service works

Extract invoice fields from supplied text, preserve currency and identifiers, and mark missing values; no OCR or accounting posting. Uses buyer-supplied information only; no external research or outbound actions. Returns a draft for human review.

## Output

Returns a structured JSON object containing a markdown table of extracted invoice fields (invoice number, seller, buyer, date, items, quantities, prices, subtotal), plus a summary, list of assumptions made, identified evidence gaps (missing fields), a flag indicating whether human review is required, provenance metadata, and a unique request ID.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "brief": {
   "type": "string",
   "maxLength": 12000,
   "minLength": 10
  },
  "audience": {
   "type": "string",
   "maxLength": 500
  },
  "source_text": {
   "type": "string",
   "maxLength": 32000,
   "minLength": 20
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "task": "invoice_text_extraction",
  "summary": "Extracted key invoice fields from the provided fictional invoice text, preserving currency and identifiers, and clearly marking the missing payment status field.",
  "artifact": "Illustrative excerpt from a synthetic provider check:\n### Extracted Invoice Fields\n\n| Field             | Value                     |\n|-------------------|---------------------------|\n| Invoice Number    | INV-1001                  |\n| Seller            | Example Supply Company    |\n| Buyer             | Example Door Company      |\n| Invoice Date      | 2026-09-01                |\n| Item Description  | entry door hardware       |\n| Quantity          | 2                         |\n| Unit Price        | USD 50                    |\n| Subtotal          ",
  "provenance": {
   "source": "buyer_supplied",
   "generated_at": "2026-09-22T14:42:47.668Z",
   "external_research_performed": false
  },
  "request_id": "example_invoice_text_extraction",
  "assumptions": [
   "The invoice number 'INV-1001' uniquely identifies the invoice.",
   "The unit price and subtotal are consistent and correctly calculated."
  ],
  "evidence_gaps": [
   "Payment status is not specified in the source text.",
   "No invoice address or contact details for seller or buyer are provided."
  ],
  "requires_human_review": true
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/tradepilotusa-invoice-text-extraction-d895ea7b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from www.tradepilotusa.com](https://www.zero.xyz/host/www.tradepilotusa.com/llms.txt)