Guide
A PDFMonkey alternative for fixed-layout PDFs, with the migration code
PDFMonkey renders HTML and CSS, with Liquid logic for loops and conditionals, into a new PDF. That's the right tool when a document needs real page design: a catalog with a grid of products, a report with charts, anything built from scratch. A lot of what teams actually generate is narrower than that. A rental application is already a PDF with named fields. An invoice is a header, two address blocks, a line item table, and a total, the same shape every time. Those don't need a template engine. They need data written into a fixed layout. Here is that migration end to end, with every request run against the live API and the actual response.
Draw the invoice from JSON, no template dashboard
Start with the invoice, since PDFMonkey's biggest use case is
probably billing. There's no template upload step. A POST to
/api/invoice with a
from block, a to block, an item list, and
a date returns a finished PDF, header, address blocks, items table,
and total laid out already.
curl -X POST https://pdfops.dev/api/invoice \
-H "Content-Type: application/json" \
-H "X-API-Key: $PDFOPS_API_KEY" \
-d '{
"from": { "name": "Birch & Co", "lines": ["44 Birch Lane", "Austin, TX 73301"] },
"to": "Jamie Oyelaran",
"invoice_number": "INV-1042",
"date": "2026-11-19",
"items": [
{ "description": "PDF form filling, 2000 docs", "quantity": 1, "unit_price": 79.00 },
{ "description": "Overage, 340 docs", "quantity": 340, "unit_price": 0.01 }
],
"notes": "Net 15"
}' \
-o invoice.pdf
That request returned a 200 and a one-page, 2,676-byte PDF with the two line items, their totals, and the note printed under them. Send the same JSON again next month for a recurring invoice and the output is byte-identical, since there's no render-time clock or random layout pass involved, so a dedup check or a cache key can hash the request and skip work it already did.
Fill the PDF you already have
Now the application form. This is the case PDFMonkey can't really touch, because the output has to be a specific PDF someone else already designed, a government form, a lease, a W-9, with fields in fixed positions. The API works from the PDF itself rather than a template description, so the first step is asking it what fields exist.
curl https://pdfops.dev/api/inspect \
-H "X-API-Key: $PDFOPS_API_KEY" \
-F "pdf=@application-form.pdf"
# {"count":8,"fields":[{"name":"applicant_name","type":"text",...},
# {"name":"email_addr",...},{"name":"dob",...},{"name":"addr_line1",...},
# {"name":"city",...},{"name":"postal_code",...},{"name":"phone_no",...},
# {"name":"notes",...}]}
Inspecting the sample application returns eight text fields:
applicant_name, email_addr,
dob, addr_line1, city,
postal_code, phone_no, and
notes. Posting those same keys with real values to
/api/fill-form writes
them into the form and hands back a filled, 11,120-byte PDF with
the original layout intact.
curl https://pdfops.dev/api/fill-form \
-H "X-API-Key: $PDFOPS_API_KEY" \
-F "pdf=@application-form.pdf" \
-F 'fields={"applicant_name":"Jamie Oyelaran","email_addr":"jamie@example.com",
"dob":"1990-04-12","addr_line1":"44 Birch Lane","city":"Austin",
"postal_code":"73301","phone_no":"512-555-0199",
"notes":"Referred by a current tenant"}' \
-o filled.pdf
Combine them into one packet
Last step, a packet. A rental application usually travels with a
cover invoice or receipt attached, so merge the two PDFs from above
into one file with /api/merge,
which takes the files as repeated multipart fields and concatenates
their pages in the order sent.
curl https://pdfops.dev/api/merge \
-H "X-API-Key: $PDFOPS_API_KEY" \
-F "pdf=@filled.pdf" \
-F "pdf=@invoice.pdf" \
-o packet.pdf
The merged file came back as a 2-page, 7,843-byte PDF, the filled
application first, the invoice second, in the order the two
pdf fields were sent. Same quota as the other two
calls, same multipart shape, no separate merge product to learn.
Try it
All three calls above ran against the live API with a key, but
none of them need one to try. Keyless access caps at 100 requests
per IP a month, enough to run this exact migration end to end. A
free key from POST /api/signup raises
that to 250 a month under the X-API-Key header and
removes nothing else, same three endpoints, same request shapes.