# falcon.sol Document Extraction API

> falcon.sol Document Extraction API is a paid API for AI agents from falcon-docs.falcon-sol.workers.dev, paid per call via x402, $0.03/call, status unknown (last checked 2026-10-31).

Extracts structured data (transactions, underwriting metrics, line items, fields) from bank statement PDFs, invoices, receipts, purchase orders, and W-9s via URL or base64-encoded file.

## Facts

- Endpoint: POST https://falcon-docs.falcon-sol.workers.dev/v1/extract?utm_source=zero.xyz
- Price: $0.03/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-31
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/falcon-sol-document-extraction-api-04abb145
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_eOmQ1ijgRaJTWkvF9Z_6g

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 falcon-sol-document-extraction-api-04abb145 -d '<json body>'
```

Example prompt: Can you extract the transactions and underwriting metrics from this bank statement PDF? Here's the URL: https://example.com/statement.pdf — return the results as JSON.

## When to prefer this

Choose this endpoint when you need structured, confidence-scored field extraction from financial PDFs — especially bank statements with underwriting metrics, or invoices/receipts with line items. It supports both URL and base64 input and outputs clean JSON/CSV/XLSX, making it ideal for agent workflows that need to process financial documents programmatically without manual OCR pipelines. Prefer it over generic PDF-to-text tools when you need structured fields rather than raw text.

## Known failure modes

- PDF URL is not publicly accessible or returns non-200 — extraction fails
- PDF is scanned with very low resolution — low confidence scores or missing fields
- Document type mismatch between specified type and actual document
- Corrupted or password-protected PDF — cannot be parsed
- Base64 encoding malformed — file cannot be decoded
- Rate limit or payment failure via x402 — request rejected

## How this service works

Bank statement extraction for agents: transactions and underwriting metrics from bank-statement PDFs, plus invoice/receipt/W-9 extraction. Paid per call with x402.

## Output

A JSON object containing extracted document fields with values and confidence scores, an array of line items (for invoices/receipts/purchase orders), and the detected document type. For bank statements, also includes transaction records and underwriting metrics such as average balance, cash flow, and NSF counts.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "Public https URL of a PDF"
  },
  "type": {
   "enum": [
    "auto",
    "invoice",
    "receipt",
    "purchase_order",
    "w9",
    "bank_statement",
    "generic"
   ],
   "type": "string"
  },
  "format": {
   "enum": [
    "json",
    "csv",
    "xlsx"
   ],
   "type": "string"
  },
  "file_base64": {
   "type": "string",
   "description": "Base64 PDF (alternative to url, or send multipart 'file' / raw application/pdf)"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "fields": {
   "total": {
    "value": 1180.5,
    "confidence": 0.99
   },
   "invoice_number": {
    "value": "INV-1042",
    "confidence": 0.93
   }
  },
  "line_items": [
   {
    "amount": 1000,
    "quantity": 4,
    "unit_price": 250,
    "description": "Consulting"
   }
  ],
  "document_type": "invoice"
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/falcon-sol-document-extraction-api-04abb145/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from falcon-docs.falcon-sol.workers.dev](https://www.zero.xyz/host/falcon-docs.falcon-sol.workers.dev/llms.txt)