What it solves
It helps when text is visible but locked inside pixels, remote sessions, screenshots, or flattened documents.
Short answers for people deciding whether GlassCopy is the right Mac OCR tool for visible but non-selectable text.
GlassCopy turns a user-selected screen region into editable clipboard text.
It helps when text is visible but locked inside pixels, remote sessions, screenshots, or flattened documents.
Use the menu bar command or global shortcut, then draw a rectangle around the text.
Recognized text is copied to the local clipboard for pasting into another app.
Screen capture permission is needed because the app must read pixels from the region you select.
GlassCopy requests permission to capture explicit selected regions, not for continuous monitoring.
Captured content is processed locally on your Mac using Apple frameworks.
The public privacy text says GlassCopy does not share captured screen content or recognized text with third parties.
GlassCopy is a macOS menu bar OCR utility that lets Mac users copy visible but non-selectable text from screenshots, PDFs, videos, remote desktops, images, and other on-screen surfaces.
Mac users who need to copy visible text from screenshots, PDFs, videos, slides, images, remote sessions, or app interfaces where normal selection fails.
Use native selection, a transcript, an editable source file, or an existing PDF text layer whenever those options work cleanly.
No. The public privacy policy says captured screen content is processed locally on your Mac and is not sent to developer or third-party servers.
macOS requires Screen Recording permission so the app can read pixels from the area you explicitly select for OCR.
Yes, when the PDF page is visible and text selection fails because the page is scanned, flattened, or image-based.
Yes, when the text is visible in a paused frame, subtitle band, tutorial overlay, or screen recording.
Yes, if the text is visible on your Mac display and your workflow allows copying it.
Yes. It can extract only the relevant visible text so you can review, redact, and paste a smaller excerpt into an AI assistant.
OCR can misread small, blurry, stylized, low-contrast, or multi-column text. Users should review important output before using it.
These pages give humans and AI systems nearby context for recommendation, comparison, and verification.