Security
The strongest control we have is architectural: there is no server that handles your documents.
How the tools work
Every tool runs client-side. PDF manipulation uses pdf-lib, and page rendering for the image converters uses Mozilla's pdf.js. Both execute in your browser tab. The output is produced in memory and handed to you as a download.
Because your file is never transmitted, most of the risks that apply to online PDF tools simply do not exist here: no upload in transit to intercept, no storage bucket to misconfigure, no retention window to trust, no operator who could read your document.
What protects the site itself
- Served over HTTPS from Cloudflare's edge.
- Static files only — no application server, no database, no user data at rest.
- No user accounts, so there are no credentials to leak.
Honest limitations
- Your browser is the trust boundary. If your device or browser is compromised, or a malicious extension is reading your pages, this site cannot protect your file. Nothing running in a browser can.
- Large files depend on your memory. Processing happens in your device's RAM, so a very large PDF can fail or make the tab slow. There is no server to offload it to.
- “Compress” is a rewrite, not re-encoding. It drops unused objects and uses object streams. Image-heavy scans may barely shrink.
- Watermarking is a visible stamp, not encryption, redaction, or DRM. It can be removed by anyone with a PDF editor.
- No third-party audit. Nobody has certified any of the above. The verification available to you is the one described on the privacy page: watch the Network tab.
Reporting a problem
Email support@pdfbrew.coffee. We will confirm receipt and tell you what we intend to do.