Sort Your Messy Downloads Folder — PDFs, Screenshots and Photos — Into Clean Files, Offline

Short answer: Your Downloads folder (and probably your Desktop) is a graveyard of PDFs, screenshots and phone photos you "saved for later." PDF Insight is a local desktop app for Mac and Windows that turns that pile into something usable. Point it at a folder, type the order you want in plain English ("group by project, newest first, OCR the screenshots"), and a local AI reads each file — including images: screenshots, scans and phone photos of documents — classifies it, sorts it, and merges everything into one clean, ordered, page-cited file per project. It all runs on your own machine; nothing is uploaded in the default local tier. It's a focused sort-and-merge tool, not an always-on file manager that reorganizes your whole disk — and we'll be clear about that line.

The folder you keep meaning to clean up

Everyone has it. The Downloads folder with 1,400 items. The Desktop where icons have stopped fitting on the screen. It's not just PDFs anymore — it's screenshots (the receipt you snapped before closing the tab, the confirmation page, the chart from a Slack thread), phone photos (a document someone handed you, a whiteboard after a meeting, a warranty card, a paper bill), and the usual landslide of invoice-final-2.pdf and scan0007.pdf.

The annoying truth is that the content is fine. You have the warranty, the receipt, the contract, the meeting notes. They're just scattered across formats and folders in zero order, which means that when you actually need "everything about the kitchen reno" or "the paperwork for this trip," you're digging through a mixed pile of PDFs and images by hand.

That's a sorting problem, not a storage problem. And it's exactly what a local sort-and-merge tool is built for.

Images are first-class here, not an afterthought

Most "merge your PDFs" tools quietly assume everything is already a clean PDF. Real life isn't like that. Half of what you need to file is an image: a screenshot, a scan, a photo you took with your phone because that was the fastest way to capture it.

PDF Insight treats those as first-class inputs. On-device OCR (Tesseract, English + French) reads the text inside a screenshot or a photographed document, so the app can tell that the blurry photo is a receipt and the screenshot is a booking confirmation — and sort them accordingly, then fold them into the merged file right alongside your PDFs.

"Take this folder, OCR all the screenshots and photos, group by project, and put each project in date order."

Reasonable expectation: OCR quality follows image quality. A crisp screenshot or a flat scan reads cleanly; a dim, angled phone photo will be rougher. It's good enough to recognize and sort the document — it isn't magic, and it isn't a typesetting engine.

Describe the order once, get one clean file

There's no syntax to learn. You say what you want the way you'd explain it to a person:

1. Point it at the folder. Downloads, Desktop, a project dump — PDFs, screenshots and photos all mixed together. Messy is fine.
2. Type the order in plain English. "Group by project. Within each, newest first. OCR the images so they're searchable."
3. Let the local AI read and sort. It reads each file, works out what it is, OCRs the images, and orders the stack the way you described.
4. Get one clean, page-cited file per project. Every section traces back to the source file and page it came from, so you can always find the original.

A few directives that work well for a chaotic folder:

"Separate this into one file per trip. Flights first, then hotels, then everything else by date."

"Group all the home-renovation paperwork together: quotes first, then invoices, then receipts and warranty photos, oldest to newest."

"Sort by date, newest first, and OCR every screenshot and photo so the text is searchable."

Why local-first is the whole point

A Downloads folder is a surprisingly intimate thing. It has your receipts, your bank PDFs, screenshots of private messages, photos of documents with your address and signature on them. The last thing you want is to upload all of that to a random web service just to tidy it up.

PDF Insight runs 100% locally by default. The AI reads and classifies on your own computer, OCR happens on-device, the merge happens on-device, and nothing is uploaded in the local tier. It works fully offline. There is an optional paid cloud speed lane — clearly labelled — but it stays off unless you deliberately turn it on. If you're the kind of person who self-hosts, keeps data off other people's computers, and reads the privacy policy before installing anything, this is built the way you'd want it built. For the longer argument, see Where Your Files Actually Live: Cloud vs Local-First and Is It Safe to Use AI on Sensitive Documents?.

Before and after, at a glance

What this is NOT (so you don't expect the wrong thing)

Let's be straight about the limits, because the wrong expectation leads to disappointment.

PDF Insight is not an always-on file manager. It does not run in the background watching your disk, and it does not silently rename or re-file everything across your whole computer. You point it at a folder when you want to, and it produces one clean, ordered, cited file. It is a deliberate sort-and-merge step, not a daemon that takes over your filesystem.

It also does not extract data into spreadsheets — it won't total your receipts, pull out vendors and amounts, or build a report. "Reading" means it understands what each file is and where it goes in the order, not what you spent or what a document concludes. And it's not cloud storage, a notes app, or a backup tool.

What it is: a focused tool that takes the messy pile of PDFs and images you already have and turns it into clean, ordered, cited files — on your own machine. If sorting and assembling is your bottleneck, it helps a lot. If you wanted it to run your whole life, that's not what this is — and we'd rather tell you now.