This quickstart guide will help you make your first API call to justextract.it in under 5 minutes. If you are looking for detailed API specifications, please refer to our reference document instead.

​
Step 1: Get an API Key

Please book a meeting and get a paid API key from our console thereafter.

Important: You will not be able to proceed without completing this step.

​
Step 2: Set your environment variable

local.env
JUSTEXTRACT_API_KEY = YOUR_API_KEY

​
Step 3: Prepare a PDF file for test extraction

To speed up experimention, we will use a publicly hosted PDF report by McKinsey to demonstrate various extraction capabilities. Alternatively, feel free to use your own PDF document hosted publicly.

Important: Google Drive links or any proprietary cloud drive URLs pointing to a PDF will not work. Only raw PDFs, e.g. those hosted on S3 will be accessible by our service for privacy reasons.

​
Step 4: Copy the code below to make a fetch request

const response = await fetch(
      "https://api.justextract.it/api/extract",
      {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
          Authorization: "Bearer JUSTEXTRACT_API_KEY",
        },
        body: JSON.stringify({
          // We will use a sample PDF report from McKinsey here
          url: "https://www.mckinsey.com/~/media/mckinsey/business%20functions/quantumblack/our%20insights/the%20state%20of%20ai/2024/the-state-of-ai-in-early-2024-final.pdf",
          filters: [
            // These filter options can be left empty for now.
			// Each filter will be applied sequentially, if it exists.
			// For example, page-based filtering will occur first.
			// Whatever remains will be filtered using a semantic query, etc.
            // Read "https://docs.justextract.it/development" for more info
            // { pages: [] },
            // { query: "" },
          ],
        }),
      }
    );

const data = await response.json();

​
Step 5: Run the code in your development environment

node api.js

A response should soon appear in your console.

​
Success Response Object

To make development as easy as possible for you, all response objects are standardized across our APIs.

After sending over a URL pointing to the document intended for extraction, you should receive an object with task_id in return:

example-response.json
{
  "task_id": "some-id"
}

For a full list of values contained in a response, please refer to our API reference.

To increase your extraction volume, you may request a custom pricing plan from us.