Turn logistics PDFs into structured JSON

After verification, the dashboard shows your full API key once (prefix ext_live_…). Store it in a secret manager — it cannot be retrieved again in full.

Auth header: Authorization: Bearer … (not X-Api-Key).

POST /api/v1/extract with a PDF and document_type. One credit per page; failures are refunded.

curl -X POST https://cyber.lerepop.lv/api/v1/extract \
  -H "Authorization: Bearer ext_live_YOUR_KEY" \
  -F "file=@invoice.pdf" \
  -F "document_type=invoice"

import requests

response = requests.post(
    "https://cyber.lerepop.lv/api/v1/extract",
    headers={"Authorization": "Bearer ext_live_YOUR_KEY"},
    files={"file": open("invoice.pdf", "rb")},
    data={"document_type": "invoice"},
)
print(response.json())

import fs from "node:fs";

const form = new FormData();
form.append("file", new Blob([fs.readFileSync("invoice.pdf")]), "invoice.pdf");
form.append("document_type", "invoice");

const response = await fetch("https://cyber.lerepop.lv/api/v1/extract", {
  method: "POST",
  headers: { Authorization: "Bearer ext_live_YOUR_KEY" },
  body: form,
});

console.log(await response.json());

$ch = curl_init("https://cyber.lerepop.lv/api/v1/extract");
curl_setopt_array($ch, [
    CURLOPT_POST => true,
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => ["Authorization: Bearer ext_live_YOUR_KEY"],
    CURLOPT_POSTFIELDS => [
        "file" => new CURLFile("invoice.pdf", "application/pdf"),
        "document_type" => "invoice",
    ],
]);
echo curl_exec($ch);

Prefer the OpenAPI explorer? API reference →

Each field returns value, confidence, and often source. Use confidence thresholds before writing into your TMS.

• Check balance: GET /api/v1/account/balance
• Buy page-credit packs via Stripe when free quota is used
• Enable async + webhooks for batch volume

No. After email verification you get 15 free page credits per month. Buy Stripe credit packs only when you need more volume.

We publish how often shipment-critical fields match a labelled lab set (shown as a %). That is not a guarantee for every customer scan. Digital invoices and PODs are strongest; hard scans and exotic customs/B/L need review more often. Run a short pilot on your own files.

Invoice, CMR, POD, delivery note, customs, bill of lading, rail waybill, plus auto-detect and combined multi-document PDFs. Pass document_type or let the pipeline classify.

We do not keep uploaded PDFs as a long-term archive for extraction cache. Field results may be cached per account for 30 days in Redis and are never shared across API accounts. See data-processing docs and the DPA for the full picture.

Digital PDFs are parsed locally (Poppler) on our servers. Scans may use Mistral OCR; optional structured stages may use OpenAI. Triggers and the full list live on /subprocessors with change notice.

Use sync POST /extract for interactive or low-volume flows. For batches, use async mode, poll request status, and optionally webhooks. Rate limit is 60 requests/minute by default.

Uploads up to 50 MB. Billable pages are capped at 20 per request — larger files return a clear validation error.

No. The interactive demo replays curated fixture PDFs and stored extraction results (with page highlights). Sign up and use the dashboard or API to run your own documents.