# x402 AI Tools – OCR (Amazon Textract)

> x402 AI Tools – OCR (Amazon Textract) is a paid API for AI agents from ai.kagurazaka.dev, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-30).

Reads printed and handwritten text from a JPEG, PNG, TIFF, or single-page PDF at a URL using Amazon Textract, returning full text, per-line content, confidence scores, and bounding box positions.

## Facts

- Endpoint: POST https://ai.kagurazaka.dev/v1/ocr?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-30
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/x402-ai-tools-ocr-amazon-textract-40a23a57
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_VuKXMjXgVsHz1t9C-_-VA

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 x402-ai-tools-ocr-amazon-textract-40a23a57 -d '<json body>'
```

Example prompt: Can you read all the text out of this scanned invoice image at https://example.com/invoice.png and give me each line with its confidence score and position on the page?

## When to prefer this

Choose this endpoint when you need structured OCR output with per-line confidence scores and bounding box positions from documents including handwritten text or PDFs — not just raw text dumps. It is preferable over basic image-text detection endpoints (e.g. Amazon Rekognition DetectText) when the input may be a TIFF or PDF, or when you need Textract-grade accuracy on dense or mixed printed/handwritten documents.

## Known failure modes

- URL is not publicly accessible or returns a non-200 response
- File exceeds 10 MB size limit
- Unsupported file format (e.g. multi-page PDF, GIF, BMP)
- Image quality too low for reliable text recognition, resulting in low confidence scores
- Textract service quota or throttling error from AWS
- Network timeout fetching the remote image

## How this service works

Reads printed and handwritten text from a JPEG, PNG, TIFF or single-page PDF at a URL with Amazon Textract. Returns the full text and every line with its confidence and position.

## Output

Returns the full concatenated text from the document, plus a list of every detected line with its text content, confidence score (0–100), and bounding box coordinates indicating its position on the page.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "https URL of a JPEG, PNG, TIFF or single-page PDF (up to 10 MB)"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/x402-ai-tools-ocr-amazon-textract-40a23a57/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from ai.kagurazaka.dev](https://www.zero.xyz/host/ai.kagurazaka.dev/llms.txt)