# I built a PDF toolkit where files never leave your browser (42 tools, public beta)

> Source: <https://dev.to/pranav_rp_c9ff6aafac0100f/i-built-a-pdf-toolkit-where-files-never-leave-your-browser-42-tools-public-beta-31k9>
> Published: 2026-11-21 10:58:31+00:00

**Key takeaway:** Running PDF editing, OCR and conversion entirely client-side is feasible by combining pdf.js, pdf-lib, and WebAssembly builds of qpdf and Tesseract inside Web Workers.

Most free PDF sites upload your file to a server. For contracts, IDs and tax forms that always bothered me, so I built Pagewright: 42 PDF tools that run entirely in your browser. Your files are processed on your own device and never uploaded.

Merge, split, rotate, crop, compress, repair, OCR, convert to and from images, Word, Excel, PowerPoint, HTML and Markdown, watermark, fill forms, redact, compare two versions, AES-256 password protect, and chain tools into reusable workflows.

React, TypeScript and Vite. pdf.js renders pages, pdf-lib edits them, and qpdf and Tesseract run as WebAssembly. Heavy work runs in Web Workers so the page stays responsive.

13 of the 42 tools are marked "limited" on their pages. Signing is a visible electronic signature, not a certificate-based digital signature. PDF/A output is prepared for archiving but not validated. Translate and summarize rely on on-device browser AI where available. The full list is in docs/known-limitations.md in the repo.

Unit tests, Playwright runs on Chromium, WebKit and Firefox, axe accessibility checks on every route, and a hardening pass on decompression limits and the CSP.

I would love feedback, especially PDFs that fail or tools that feel clumsy.
