Your files
stay here.

When you open a file, the browser gives this page read access to that file only. Tools read it into memory as an ArrayBuffer or Blob and process it with JavaScript, WebAssembly, Canvas and Web Workers running on your device.

Results are produced as local Blobs and saved through your browser’s normal download mechanism. There are no API routes, no server functions, no ads and no third-party trackers in this application.

Heavy work (image filters, CSV profiling, ZIP, PDF assembly, regular expressions) runs in Web Workers — background threads in this same tab — so the interface stays responsive.

The only network requests are for this application’s own code and styles, from the server that hosts it — the same as any website. After the first visit those files are cached on your device so the tools keep working offline.

To prevent content from making requests on its own, untrusted markup is sanitized: SVG files have scripts and external references removed, and the Markdown preview blocks remote images unless you explicitly allow them.

Using IndexedDB in this browser profile:

• Favorite and recently used tools
• Recent files: name, size, type and a small thumbnail (turn off in Settings)
• Copies of recent files — only if you enable “Keep copies” in Settings
• Preferences (such as sidebar state); the theme choice is kept in local storage
• Projects you explicitly save (for example in the Photo Editor), including their source images

Clearing site data in your browser, or using the button below, removes everything.

Checking…

Some features depend on what this browser supports. Tools adapt automatically.

• Web Workers — background processing
• OffscreenCanvas — image processing off the main thread
• Web Crypto — SHA hashing and secure random UUIDs
• WebP encoding — export images as WebP
• AVIF encoding — export images as AVIF
• File System Access — native open/save dialogs (optional)

Browsers limit memory per tab. Very large images, PDFs or datasets may exceed that limit; when this happens the tool tells you and nothing is lost. Some formats (such as HEIC images or encrypted PDFs) cannot be decoded by every browser.

JWT signatures can be verified for HMAC secrets only. Client-side PDF compression cannot match desktop software in every case, and scanned PDFs can’t be turned into text because OCR isn’t available offline.

To know how many people use Jellykit, we count page visits with our own self-hosted, open-source Umami analytics. It records the page address, the referring site, your country, browser and device type — without cookies and without storing your IP address.

It never sees your files or anything you type into a tool: those stay in this browser, as described above.

Jellykit is free, with no ads and no third-party tracking. If it helps you, you can buy me a coffee. That link is an ordinary link to Buy Me a Coffee — nothing from their site is loaded unless you click it.