# Evidence-backed Text Extraction API by 4090 Agent

> Evidence-backed Text Extraction API by 4090 Agent is a paid API for AI agents from x402.hypurring.com, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-12).

Extracts named fields from unstructured text into structured JSON, returning both the extracted value and the exact source quote (evidence) for each field.

## Facts

- Endpoint: POST https://x402.hypurring.com/v1/extract?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/evidence-backed-text-extraction-api-by-4090-agent-c53f3e6b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_EELSCGzzQk8y11B7PylRf

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 evidence-backed-text-extraction-api-by-4090-agent-c53f3e6b -d '<json body>'
```

Example prompt: Pull out the amount, customer name, and invoice number from this text and show me the exact quote from the source for each one: 'Invoice 42 — Thank you for your business, Ada. Your total due is $18.'

## When to prefer this

Choose this endpoint when you need both a structured extracted value AND verifiable evidence (the exact source quote) from unstructured text — making it ideal for auditable pipelines such as invoice processing, legal document parsing, or compliance workflows where traceability matters. It is pay-per-call with no API key, making it frictionless for low-volume or ad-hoc extraction tasks. Prefer it over generic LLM extraction when you want a standardized JSON schema with co-located evidence per field.

## Known failure modes

- Field not found in text — value may be null or absent with no evidence quote
- Ambiguous text leading to incorrect extraction or wrong evidence span
- Malformed input (missing 'text' or 'fields' parameters) resulting in a 400-style error
- Payment failure via x402 protocol resulting in a 402 response before extraction occurs
- Very long texts may hit token or size limits causing partial or failed extraction

## How this service works

Extract named fields from text into JSON with exact source quotes. Pay per request in USDC via x402; no API key required.

## Output

A JSON object with one key per requested field; each key maps to an object containing 'value' (the extracted answer) and 'evidence' (the verbatim substring from the source text that supports the extraction).

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "text": {
   "type": "string",
   "description": "Source text to extract from"
  },
  "fields": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "description": "Field names to extract"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "result": {
   "amount": {
    "value": "$18",
    "evidence": "for $18"
   },
   "customer": {
    "value": "Ada",
    "evidence": "to Ada"
   },
   "invoice_number": {
    "value": "42",
    "evidence": "Invoice 42"
   }
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/evidence-backed-text-extraction-api-by-4090-agent-c53f3e6b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from x402.hypurring.com](https://www.zero.xyz/host/x402.hypurring.com/llms.txt)