Home›For Developers

imisspdf for Developers

Build privacy-first PDF features into your own app — using the same engine that powers imisspdf.com.

npm package

`@imisspdf/pdf-utils`

Tiny, well-typed wrapper around pdf-lib for the 8 most common PDF operations. Works identically in the browser, Node 18+, Bun, and Deno. MIT licensed.

Install

npm install @imisspdf/pdf-utils

Why this exists

pdf-lib is excellent — but for the 5 operations 90% of apps actually need (merge, split, rotate, count pages, watermark), the code is verbose and the same boilerplate gets re-copied across projects. This package is the boring missing layer:

Concise API — mergePdfs([buf1, buf2]) instead of 20 lines
Well-typed — full TypeScript types, JSDoc on every public function
Universal — identical API in browser (Vite, Webpack, ESBuild) and Node 18+
No network — not a wrapper around a SaaS — does everything locally with WebAssembly + pdf-lib
Minimal surface — 8 functions, no DI, no plugins, no config
MIT licensed — use in commercial products with no obligations

Quick examples

1. Merge multiple PDFs into one

import { mergePdfs } from '@imisspdf/pdf-utils';

// In the browser, from <input type="file" multiple>:
const files = Array.from(input.files!);
const buffers = await Promise.all(files.map((f) => f.arrayBuffer()));
const merged = await mergePdfs(buffers);

// merged is a Uint8Array — save it:
const blob = new Blob([merged], { type: 'application/pdf' });
const url = URL.createObjectURL(blob);
window.open(url);

2. Split a PDF by page range

import { splitPdf } from '@imisspdf/pdf-utils';

const pdf = await readFile('./report.pdf');
const subset = await splitPdf(pdf, [
  { from: 1, to: 3 },
  { from: 7, to: 10 },
]);
await writeFile('./excerpt.pdf', subset);

3. Add a text watermark

import { addTextWatermark } from '@imisspdf/pdf-utils';

const stamped = await addTextWatermark(pdf, 'CONFIDENTIAL', {
  opacity: 0.2,
  fontSize: 60,
  angle: -30,
});

Full API surface

Eight functions covering the operations 90% of apps need. Full types in src/index.ts.

mergePdfs(buffers) — concatenate PDFs in array order
splitPdf(buffer, ranges) — extract page ranges
rotatePdf(buffer, opts) — rotate all or selected pages
getPageCount(buffer) — count pages without rendering
addTextWatermark(buffer, text, opts?) — stamp text on every page
getMetadata(buffer) — extract title, author, creator, dates
removeMetadata(buffer) — strip metadata for privacy
isPdfEncrypted(buffer) — pre-flight check

Browser vs Node — what works where

Every function is identical in both runtimes. The library uses standard ArrayBuffer and Uint8Array types, no Node-specific or DOM-specific APIs.

Browser (Chrome, Safari, Firefox, Edge — modern) — works via WebAssembly, ~80 KB gzipped
Node 18+ — use fs/promises for file I/O, then pass ArrayBuffer
Bun & Deno — tested, works
Edge runtimes (Cloudflare Workers, Vercel Edge) — works (no Node-specific APIs)

Privacy posture

This is a client-side library. It does not make any network requests. It does not collect telemetry. It does not include analytics. The only dependency is pdf-lib, which is similarly self-contained.

If you're building a privacy-first product, your dependency tree stays clean. Audit with npm ls --all — you should see only pdf-lib.

Roadmap

Planned for 0.2.x (issues welcome with use-case votes):

compressPdf — image re-sampling and stream re-encoding
flattenForms — bake interactive fields into the visible layer
addPageNumbers — with templates
addImageWatermark — PNG/JPG stamps with positioning
extractTextLayer — for indexing / search use cases

Why not a REST API?

We considered offering a hosted REST API for server-to-server PDF processing. We decided not to ship one because it would contradict the architectural promise of imisspdf.com — that your files never leave your machine. A REST API necessarily uploads files to our servers. If you want that, Adobe, PDFKit.io, and ConvertAPI already do it well. Use them when server-side processing fits your use case.

If you want privacy-by-default client-side PDF processing, @imisspdf/pdf-utils is for you.

License & contributing

MIT — use in commercial products with no obligations.

We accept PRs. The library is small and the bar is "well-typed, no network calls, no breaking the public API". Issues and feature requests welcome. We prioritise additions that real apps would use.

View on GitHub →

Tools

Solutions

Company

Product