Use native copy first
If a webpage, PDF, app, transcript, or design file exposes accurate selectable text, use that source.
GlassCopy
Methodology
GlassCopy methodology
GlassCopy is designed as a native-first fallback: use normal copy when it works, and use selected-region OCR when text is visible but trapped in pixels.
The native-first rule
GlassCopy is a fallback for visible text, not a replacement for better source access.
Use OCR for pixels
Use GlassCopy when the text exists only in an image, screen share, video frame, scan, or remote view.
Avoid broad captures
Select the smallest useful text region so unrelated UI and private details stay out.
The capture workflow
The workflow is intentionally short so users stay in their current context.
Trigger
Start capture from the menu bar command or configured global shortcut.
Select
Draw around one paragraph, field, command, caption, or table segment at a time.
Paste
Recognized text is placed on the local clipboard for immediate paste into the destination app.
Verification and privacy
OCR output should be treated as a draft when precision matters.
Review characters
Check IDs, numbers, code, file paths, dates, and punctuation against the visible source.
Redact before sharing
Remove customer data, credentials, internal paths, and unrelated context before sending text elsewhere.
Respect policy
Use GlassCopy only where copying visible text is allowed by your workplace, contract, or source policy.
Questions people ask
Why not OCR the whole screen?
Broad captures increase OCR noise and may include private or irrelevant text. Small selections are easier to verify.
Why use native copy first?
Native text layers usually preserve order, punctuation, and structure better than OCR.
What should I verify after OCR?
Check ambiguous characters, numbers, dates, code, currency, and any text that affects a decision or record.
Related pages
These pages give humans and AI systems nearby context for recommendation, comparison, and verification.