What's inside
Features & How It Works
224 of 238 tools process your files locally in the browser tab. This page documents the engines behind them, the guarantees you get, and the limits we choose to state plainly rather than paper over.
Three steps, no upload
- 1
Pick a tool
Search by task or file format, or browse 13 categories. Every tool page tells you exactly what it accepts and produces.
- 2
Drop your files
Drag & drop, browse, or capture with your camera. Files are validated by size, extension and magic bytes before anything runs.
- 3
Download the result
Watch real progress, cancel if you change your mind, then save the output — or a ZIP for batch jobs.
Capability tour
Full page surgery
Merge, split, extract, delete, duplicate, rotate, reorder, interleave and rebuild documents page by page from a live thumbnail grid.
Real compression
Pages are re-rendered at your chosen DPI and JPEG quality, then rebuilt — you see the before/after size, not a promise.
Editing & markup
Add text, images, shapes, freehand ink, highlights, watermarks, stamps, page numbers, headers, footers and Bates numbers.
Security you can verify
Password encryption and permissions, permanent redaction, sanitizing, annotation stripping, and drawn or typed signatures.
OCR & extraction
Tesseract reads scanned pages (English & Hindi), producing searchable PDFs, text, CSV tables, images, bookmarks and metadata.
Image pipeline
Resize with 50+ presets, crop, rotate, flip, filter, watermark, blur, sharpen, upscale, palette-pick and EXIF tools — in bulk.
Capture & creation
Scan pages with your phone camera, build blank pages, fillable forms, QR codes, barcodes and PDFs from text, HTML or Markdown.
Batch everywhere
Up to 50 files per job with per-file progress, cancellation and a single ZIP download when a tool produces many outputs.
13 tool categories
The processing engine
No proprietary server, no hidden queue. These are the libraries doing the work inside your browser, loaded once and cached.
| Technology | What it does here | Runs |
|---|---|---|
| @cantoo/pdf-lib | Creates, edits and rewrites PDF structure — pages, forms, encryption, annotations, metadata. | Client |
| PDF.js | Renders and reads PDF content: page previews, positioned text, links, fonts, table heuristics. | Client |
| Canvas 2D + OffscreenCanvas | Image decoding, resizing, cropping, filters, watermarking and encoding to JPG/PNG/WEBP. | Client |
| Tesseract.js (WASM) | Optical character recognition for scanned PDFs and photos, with an invisible text layer option. | Client |
| exifreader | Reads and rewrites EXIF/IPTC metadata, GPS stripping and camera details. | Client |
| heic-to / utif2 | Decodes HEIC/HEIF and TIFF files that browsers cannot open natively. | Client |
| jsqr + jsbarcode | QR code generation and decoding; 1D barcode rendering. | Client |
| fflate | ZIP bundling for multi-output jobs, packed in the browser. | Client |
| WebCrypto (crypto.subtle) | SHA-1/256/384/512 file fingerprints and verification. | Client |
Privacy guarantees
No upload step exists
File inputs are read with the FileReader/File interface and kept in tab memory. There is no endpoint that accepts document bytes.
No accounts, no tracking of content
Nothing is stored server-side about your files. Theme, recent tools and favourites live in your own localStorage.
Open inspection
The entire processing pipeline is client JavaScript you can read in DevTools — no black box.
Safety limits, not surveillance
A 100 MB per-file and 50-file per-batch ceiling exists to protect your own browser memory, enforced locally.
Details in the Privacy Policy and the Security page.
What we do not fake
A tool that silently returns your original file, or converts a PDF into an empty Word document, is worse than no tool at all. These are the operations we deliberately label Partial or Coming soon, and why.
Office documents (Word, Excel, PowerPoint)
Faithful DOCX/XLSX/PPTX → PDF rendering needs a desktop layout engine such as LibreOffice on a server. We mark those tools “Coming soon” instead of returning a blank or unchanged PDF. In the meantime PDF → Word/CSV/TXT works by text extraction.
PDF linearization / “Optimize for web”
Byte-wise linearization needs native tooling (qpdf). Use Compress PDF to reduce size today.
Machine-learning background removal & upscaling
Our in-browser removal is color/edge based — excellent on studio-solid backgrounds. True segmentation and super-resolution need a GPU model, which we will only add as a clearly-labelled server feature.
Digital signature trust chains
We can place signatures and list existing signature fields with signer metadata. Validating certificates against system trust lists is outside what a browser can do.
Ready when you are
Pick from 238 tools, or search for the exact job in the header at any time.