▌ GitHub radar

mac-ocr: free local text recognition on your Mac

A tiny command-line tool that pulls text out of any image or PDF using Apple's built-in Vision engine — entirely on your machine, nothing uploaded.

01privatenumber/mac-ocr 379Swift

mac-ocr is a command-line tool that extracts text from images and PDFs using Apple's own Vision framework — the same recognition engine baked into macOS. Everything runs locally and nothing is uploaded, and it can also rebuild a scanned PDF into a searchable one with a real selectable text layer. It works both as a CLI and as a Node.js library, and ships an agent skill so AI assistants can read documents through it.

Why a vibe-coder should care

If you've ever screenshotted a receipt, a slide or a page of a book and wished you could just copy the text, this does it for free without burning cloud vision tokens. And if you build with AI, it's a cheap way to let an assistant read documents straight off your disk.

Open on GitHub →