No. The files you choose are read by JavaScript into browser memory, processed there by pdf-lib and pdf.js, and saved back to your downloads. There is no upload endpoint — you can load the page, go offline and everything except the first load still works.
Free PDF Tool — 100% In-Browser
PDF Toolkit — Merge, Split, Compress, Fill Forms
The everyday PDF jobs in one page: merge files, keep only the pages you need, rotate what scanned sideways, compress that 40 MB monster before e-mailing it, fill a form without printing it, stamp a watermark, drop page numbers, or turn photos into a single PDF. Your files are opened, edited and saved by JavaScript in your own browser — nothing is ever uploaded.
The PDF Jobs That Eat Your Afternoon
Merging three quotes into one file, cutting a 90-page report down to the 6 pages the client asked for, getting a 40 MB scan under an e-mail limit, and filling a two-line form that would otherwise mean print, sign, scan. Each of these is one tab here: pick the files, press the button, get the result — and because everything runs in your browser, contracts, invoices and IDs are never sent to any server.
The heavy lifting is done by two trusted open-source libraries, pdf-lib and pdf.js, loaded from a CDN and executed on your machine. The page itself is a plain PHP include on systron.net.
How It Works
1. Pick files
Choose the PDFs (or images) from your device — they open in browser memory only.
2. Set options
Page ranges, rotation, quality, watermark text — whatever the job needs.
3. Run it
The button processes everything locally with pdf-lib. Large files just take a moment.
4. Download
The edited PDF saves straight to your downloads. Your original files are untouched.
PDF Toolkit
Add text to a PDF
PDF coordinates start at the bottom-left corner; 595 x 842 pt is A4 portrait. This stamps new text on top of the page — it does not edit existing text.
Compress a PDF
Fast mode never loses quality but shrinks modestly. Strong mode converts each page to an image (text stops being selectable) — it works wonders on scanned documents. The result shows honest before/after sizes and tells you if the file grew instead.
Fill PDF forms
—
Press Load: the page renders as a preview with a numbered box over every detected field — click a box to jump straight to its input in the list below. Scanned, flattened and XFA forms have no detectable fields; for those, use the PDF Visual Editor tab: it works on any PDF.
Combine images into a PDF
JPG and PNG only (that is what the PDF format stores natively). The result lists any skipped files. For documents shot as photos, A4 fit usually looks best.
PDF info & metadata editor
Load reads the current values into the fields; edit, then Save. Strip empties everything — useful before sending a file outside the company.
Merge PDFs
On most browsers files arrive in the order you Ctrl/Cmd-clicked them. If the order is off, check the page count in the result and run Organize & Split afterwards.
Organize pages & split
Organize builds a new PDF from the kept ranges with the chosen rotation. Split downloads every page as its own file — browsers may ask permission for multiple downloads, and 100-page PDFs mean 100 files.
Watermark & page numbers
The watermark runs diagonally across the middle of every page at the chosen opacity. Both operations can run in one pass.
PDF Visual Editor — text, whiteout & images on any PDF
—
Five tools, fully in your browser, no upload: Edit Text re-opens existing text with matched font; Whiteout rubs away old text; Text and Image place new content (image size = the box you drag). Everything you place is selected immediately: left-click an item to select it, drag its body to move (left or right mouse button), drag a corner handle to resize, double-click placed text to re-edit. Text supports multiple lines (Enter in the box) and left / center / right alignment to your click point. The preview needs pdf.js; if it is blocked, the Add Text tab places text by position without a preview.
Files are opened with the browser FileReader API and processed in memory by pdf-lib. Nothing is uploaded — works offline once the page is loaded.
Toolkit Behavior, Explained
| Decision | Default in the tool | Why |
|---|---|---|
| Compression — fast mode | Lossless re-save with object streams | Removes structural waste without touching content; text stays selectable, but already-optimized PDFs barely shrink. |
| Compression — strong mode | JPEG at quality 0.6, scale 75% | Scanned pages are images anyway; re-encoding them cuts the most bytes. Trade-off: text is no longer selectable or searchable. |
| Split output | One file per page, named page-1.pdf, page-2.pdf … | No zip library is pulled in — keeping the page light matters more than a single archive download. |
| Form filling | Flatten on by default | Locked, predictable output — recipients cannot accidentally edit your entries. Uncheck to keep fields editable. |
| Page range syntax | 1-3,7,10-12 (1-based) | Reads the way people say it out loud; duplicates and out-of-range numbers are silently dropped. |
| Rotation | Added to existing page rotation, normalized to 0/90/180/270 | A page already rotated 90 that you rotate 90 again must end at 180, not back at 90. |
| Images to PDF | JPG/PNG only, A4 fit | Those are the two formats PDF stores natively — anything else would mean a full image decoder for little benefit. |
| Metadata strip | Empties Title, Author, Subject, Keywords, Creator, Producer | The six fields office tools read and index — stripping them keeps internal names and software out of shared files. |
| Password protection | Not offered | The browser library used here cannot encrypt PDFs, and shipping a fake button would be worse than an honest omission — use qpdf or your PDF reader for that. |
| Editing existing text | Edit Text tool (matched font), Whiteout + retype for scans | The text layer is read with pdf.js; the original line is covered and re-opened in place at the same size and font family, so retyped text blends in. Scans carry no text layer, so the cover-and-retype route stays available. |
| Form field locations | Numbered boxes drawn on a live page preview | A form is unreadable as a list of internal field names — the preview shows each field where it sits on the page, and clicking a box jumps to its input. |
PDF Toolkit — FAQ
Are my PDFs uploaded to a server?
How much smaller will compression actually get my file?
Honestly: it depends. Fast mode typically trims 5–20% and never changes content. Strong mode re-encodes every page as JPEG — scanned documents often shrink to 10–30% of the original, while clean text PDFs can end up slightly bigger (the tool warns you when that happens instead of hiding it). Already-compressed files are close to their floor; no tool can squeeze them much further without visible quality loss.
Does compression lose quality?
Fast mode: no, it is lossless. Strong mode: yes, by design — pages become images at the JPEG quality and scale you set. At quality 0.6 and 75% scale, scans look fine on screen and in print; text is no longer selectable or searchable. If you need to OCR or search the result, use fast mode instead.
Why does the Form Filler find no fields in my form?
The tool fills real AcroForm fields — the boxes in PDFs that were created as fillable, and it shows a numbered box over each one on the page preview so you can see what goes where. A scanned form, a photo of a form, or a PDF that was already flattened has no fields left; there is nothing to detect. For those, use the PDF Visual Editor tab to cover the old text and stamp the values at the right spots instead.
Can I edit or delete existing text in a PDF?
Yes — the PDF Visual Editor's Edit Text tool is the closest a browser tool can honestly get to live editing: it reads the text layer, covers the original line, and re-opens it as an editable box at the same spot with the matched font size and family (sans / serif / mono). You type the new words and press Enter. For scans and flattened PDFs that carry no text layer, use the two-step equivalent: the Whiteout tool rubs over the old text with a cover rectangle in the page background color, then the Text tool types the new value on top.
Can I password-protect a PDF here?
No, and that is deliberate. The in-browser PDF library this tool uses does not support encryption, and a button that appears to protect a file without actually encrypting it would be dangerous. To encrypt locally, use qpdf (qpdf --encrypt) or the password option in your PDF reader — both keep the file on your machine too.
Why did my merged PDF come out in the wrong order?
Files are merged in the order the browser reports your selection, which on most systems is the click order — but a few file pickers sort by name regardless. Check the page count in the result card; if the order is wrong, merge again in a different sequence, or fix it afterwards with Organize & Split using page ranges.
Which file types can I turn into a PDF?
JPG and PNG. Those are the only two image formats the PDF format embeds natively, so they convert without re-encoding and at full quality. HEIC, WebP, TIFF or GIF need converting first (any image viewer can do that), then they merge here in seconds.
My PDF is very large. Any limits?
The limit is your device memory, since everything happens in the browser — roughly 200 MB is where mobile devices start struggling, while a desktop handles several hundred. There is no server-side limit because there is no server involved. For strong-mode compression of big scans, expect a few seconds per 100 pages on a laptop.
My form has no fillable fields — how do I fill it anyway?
Use the PDF Visual Editor tab. Many real-world PDFs — scans, flattened documents, and XFA forms common with banks and government sites — carry no AcroForm layer, so no tool can auto-detect their fields. The PDF Visual Editor sidesteps the problem entirely: it renders the page as a preview, you click exactly where each value belongs, and the text is drawn into the real PDF at those coordinates. The Form Filler stays the better choice when fields do exist (typed values land in the actual fields and can be flattened); the editor is the universal fallback that works on anything a PDF viewer can display.