HUGEpdf.io / quickstart

Run your first PDF in one agent command.

1. Give This To Your Agent

Process this PDF with https://hugepdf.io/llms.txt
[PDF URL or local file]
[what you want extracted]

Example:

Process this PDF with https://hugepdf.io/llms.txt
https://example.com/catalog.pdf
Extract every product into JSON with sku, name, description, and price.

2. What The Agent Will Do

  1. Submit the PDF.
  2. Receive a page count, job ID, and output token.
  3. Optionally preview page 1.
  4. Activate free processing using the public preview token in llms.txt.
  5. Monitor progress.
  6. Download results.json and page-level artifacts.
  7. Retry failed pages if needed.

3. Price

Free during the public preview, up to 100 pages per PDF. No account or wallet required.
Paid access and pricing will be announced separately. Please do not send payments.

Usage is subject to per-IP and server capacity limits. If the service is busy, retry as instructed.

4. Fast Path For Software Agents

Save job_id, output_token, status_url, and outputs_manifest_url from the submission response. Then POST to the returned activation_url with the public preview token. Full instructions are in llms.txt.

Free activation window: 600 seconds after submission. This is not a payment deadline; unactivated jobs expire.

5. Preview Capability

Not sure about quality? The agent can request a first-page sample before activating free processing: 

GET /api/jobs/{job_id}/preview?token={output_token}

The sample is optional. The complete document is processed after free activation.

6. Human Status Page

Every job also has a browser status page: https://hugepdf.io/jobs/{job_id}

7. Machine Spec

Agents should follow https://hugepdf.io/llms.txt. Humans can inspect https://hugepdf.io/llms.