Off in the app until you switch them on. Never in the installer, never bundled software.
✓No account
Install and open a file. There is nothing to sign up for.
✓Your files stay on your PC
Background removal and text recognition run on your computer. Nothing is uploaded.
What it does today
Phase 1 is finished: open, view, navigate and search. Everything below works now.
Images open instantly
JPEG, PNG, WebP, BMP, GIF and TIFF. Wheel to zoom, drag to pan, fit or actual size, rotate, full screen. Photos are turned upright automatically from their EXIF orientation.
Walk a folder with the arrow keys
The folder is read once, then sorted the way a person reads it: IMG2 comes before IMG10. A filmstrip of cached thumbnails sits under the image.
Real PDF viewing
Continuous pages, a page thumbnail sidebar, fit-width and fit-page. The text layer is real text: select it, copy it, and find in document with highlighted matches.
It tells you what a file is
Dimensions, colour depth, camera, lens, exposure and the full EXIF table for images; page count, size, author and version for PDFs — plus whether a PDF is text or a scan.
Korean and English, light and dark
The interface follows your system theme and switches language without a restart.
One bad file fails alone
A damaged image or an unparsable PDF produces one clear sentence. It never takes down the program, and it never shows you a stack trace.
Formats
Images
JPEGPNGWebPGIFBMPTIFFSVGPSDJXLICODXF
Documents
PDFHEICAVIFRAW
Video
MP4MOVWebM
Local first, by construction
This is a property of the build, not just a promise in a policy. Your files are processed on your computer and never uploaded, and the page the interface runs in cannot load anything from a remote address. The program goes online only when you ask it to - for an optional AI model, or sponsored pictures you switch on - and even then sends no file, file name or personal data. A file becomes readable only after you point at it: through the dialog, a drop, or opening it with UFDI from Explorer.
No telemetry, no analytics, no crash reporting in the program
Settings, thumbnails and logs stay in your own user folder
Release logs record warnings and errors, never file contents
Built in order. A phase ships when it works in real use, not when it compiles.
Phase 1Shell, image viewer, PDF viewerDone
Phase 2Crop, resize, rotate, export, undoNext
Phase 3Conversion and batch processingPlanned
Phase 4Merge, split and reorder PDF pagesPlanned
Phase 5OCR in Korean and EnglishPlanned
Phase 6Searchable PDF from a scanPlanned
Questions
Is it really free?
Yes. There is no paid tier and no subscription. The program shows sponsored pictures only if you switch them on, and the website may show ads. The source is public under the MIT licence, so you can check that for yourself.
Will an update install something I did not ask for?
No. That behaviour is the reason this program exists. The installer contains the viewer and nothing else.
Does it send my files anywhere?
No. Your files - including photos you remove a background from or read text in - are processed on your computer and never uploaded. The program goes online only for an optional AI model you choose to download, or for sponsored pictures you switch on, and sends no file or personal data. You can confirm it by reading the source or by watching your firewall.
Can it do OCR or editing yet?
Yes. Cropping, resizing, rotating and exporting, batch conversion, one-click background removal and text recognition in photos are built in and run on your computer. Text recognition in scanned PDFs and PDF page tools come next, as the roadmap above shows.
