API Overview

The Papermill API generates PDFs from Press documents, markdown, JSON, and CSV over plain HTTPS. Every endpoint lives under a single base URL:

https://api.papermill.io

TIP

If you have not made a request to Papermill before, the Quickstart walks through your first render end to end. This API section is for looking up exact behaviour once you are building an integration.

• The main Papermill endpoints: render a PDF, validate a payload, verify an API key, and read account usage.
• Errors — the error response shape and the codes you can branch on.

Authentication

Every endpoint requires an API key. Create one from Settings → API keys in the Papermill app.

We support two ways of receiving your API key:

Header	Example
Authorization	Bearer pap_live_abc…
x-api-key	pap_live_abc…

shell

curl https://api.papermill.io/v2/authcheck \
  -H "Authorization: Bearer $PAPERMILL_API_KEY"

python

import os
import requests

response = requests.get(
    "https://api.papermill.io/v2/authcheck",
    headers={"Authorization": f"Bearer {os.environ['PAPERMILL_API_KEY']}"},
)
print(response.json())

javascript

const response = await fetch('https://api.papermill.io/v2/authcheck', {
  headers: { Authorization: `Bearer ${process.env.PAPERMILL_API_KEY}` },
})
console.log(await response.json())

Treat the key as a secret: it carries the full permissions of your account, so keep it server-side rather than in browser or mobile code. If a key is exposed, revoke it from the same settings page and issue a new one — Papermill rejects requests using a revoked key with the ERR_KEY_REVOKED code.

Content types

The POST /v2/pdf and POST /v2/validate endpoints that accept a Press payload decide how to read your request body from the Content-Type header:

Content-Type	Body is treated as
text/xml, application/xml, text/plain	A complete Press document.
application/json	Data merged into a template's <data> section.
text/csv	Tabular data merged into a template's <data> section.
text/markdown	Markdown placed into one of the template's flows.

Papermill rejects any other type with a 415 naming the type you sent and the ones it accepts.

The last three carry data rather than a whole document, so they need template_id to say which template to merge into. Without it, Papermill rejects the request with a message, e.g., template_id query parameter is required when posting JSON data. See Template & Payload for how merging works — a payload should carry only the values that change per render, since the template supplies the defaults.

Machine-readable specification

An OpenAPI 3.1 description of the endpoints on this page is published at /openapi.json. Import it into Postman or Insomnia, or generate a client from it:

curl -O https://docs.papermill.io/openapi.json

Beyond the HTTP API

If you are connecting an AI agent rather than writing an integration by hand, Papermill also runs a Model Context Protocol server that exposes template authoring and rendering as tools. See MCP Server.