PDFCraft is a free set of PDF tools that run entirely inside your web browser. Merging, splitting, compressing, signing, protecting and editing all happen on your own device, which means your documents are never uploaded, never queued on someone else's server, and never held anywhere you cannot reach. You can verify that claim yourself: load the site, disconnect from the network, and the tools keep working.
🛡️ Zero File Collection: Your documents never touch our servers. Every calculation, merge, crop, and signature rendering happens entirely inside your local web browser using JavaScript and standard HTML5 APIs.
Our Philosophy
Most online PDF tools require you to upload your sensitive documents—invoices, tax forms, contracts, and legal IDs—to their backend cloud servers. Once uploaded, you lose control over who views, logs, or stores your files. We believe this is an unacceptable privacy risk.
PDFCraft was built to show that serious file processing can happen on the client side. Because every operation runs in your browser, there is no upload step and therefore no third-party processor holding your documents — which is usually the part of a workflow that creates difficulty under GDPR or CCPA. That is a meaningful reduction in risk, but it is not a compliance certification, and nothing here is legal advice.
Core Technology
PDFCraft is a static website. There is no application server, no database, and no backend that touches your files. Everything that happens to a document happens in JavaScript on your own machine, using a small set of open-source libraries pinned to exact versions:
- pdf-lib 1.17.1 — document assembly. It builds the output file: copying pages, embedding fonts and images, and writing the final PDF structure.
- PDF.js 3.4.120 (Mozilla) — rendering. It draws pages to a canvas so you can see what you are working on, and it is the engine behind page thumbnails and image export.
- JSZip 3.10.1 — packaging. When one PDF becomes many images, the archive is assembled in the browser rather than on a server.
- HTML5 Canvas — the signature pad, the text placement layer, and the rasterising step used by the compressor.
Those libraries are served from this domain rather than from a public CDN. That is a deliberate choice with two consequences worth stating: the code that runs is the exact version that was reviewed and cannot be changed underneath you by a third party, and the tools keep working when you are offline, because nothing needs to be fetched at the moment you use them.
Who builds this
PDFCraft is built and maintained by one independent developer. There is no company behind it, no investors, and no sales team. That shapes what the site is and is not: a small number of tools that try to do their job properly, rather than a broad catalogue of features that each work approximately.
It also means the roadmap is driven by what people actually report. If a tool mishandles your document, telling us is genuinely the fastest route to it being fixed.
How PDFCraft is paid for
The tools are free. There is no paid tier, no account to create, no daily limit on how many files you can process, and no watermark stamped on anything you download. Because the processing happens on your device rather than on rented servers, the cost of running the site does not grow with the number of people using it — which is what makes giving it away sustainable.
Hosting is currently paid out of pocket, and voluntary donations help offset it. There is no advertising on the site today. If that ever changes, ads will sit around the page rather than inside the workspace, and will never be something you have to interact with in order to use a feature.
What PDFCraft deliberately does not do
A short list of things the site could plausibly offer and does not, because they cannot be done honestly in a browser:
- No accounts and no file storage. There is nothing to sign up for and no stored document to delete later, because none is ever kept.
- No optical character recognition. A scanned page stays an image here. Tools that promise to make scans searchable are doing real work that this site does not attempt.
- No PDF-to-Word or PDF-to-Excel conversion. Reconstructing an editable document from a finished PDF is a hard problem that these libraries do not solve, so it is not offered rather than offered badly.
- No inspection of your documents. Analytics record that a page was visited. They cannot record what was in your file, because the file never reaches anywhere that could look at it.
Where in-browser processing runs out
Local processing buys privacy, and it costs capability. Being straight about the trade matters more than marketing around it:
- Your device is the limit. There is no server to hand a large job to. A file big enough to exhaust your browser tab fails rather than queueing, and an older phone will manage less than a desktop.
- Compression is lossy. The compressor works by re-rendering each page as an image, so text stops being selectable or searchable. It is explained in full on that tool's page, and you should keep your original.
- Text editing is constrained by the file itself. PDFs usually embed only the characters they actually used, so the letters available to type are limited by what the document already contains.
- Drawing over content does not remove it. A signature or a box is painted on top; the text underneath is still there. Removing content requires the tools that rebuild the document.
Get in Touch
We are constantly seeking feedback to improve our tools. If you have feature requests, bug reports, or partnership inquiries, please visit our Contact Us page or drop us an email at support@pdf-craft.online.