Snaptext — Screenshot To

SnapText — Screenshot to Text

Stop retyping what's already on your screen.

SnapText lets you drag a selection over any part of any webpage — a table, a code block, a scanned image, a UI screenshot, a PDF preview — and instantly extracts every word from it. No copy-pasting. No manual transcription. Just select and read.

── How it works ──────────────────────────

1. Click the SnapText icon in your toolbar
2. Click "Capture Region" — your popup closes and a crosshair appears
3. Drag to select any area on the page
4. Text is extracted and copied to your clipboard automatically

That's it. No forms. No upload dialogs. No waiting.

── Two OCR engines. You choose. ──────────

▸ Claude Vision (recommended)
Powered by Anthropic's Claude — the most accurate OCR available. Handles complex layouts, mixed fonts, handwriting, math, tables, and code. Requires a free Anthropic API key.

▸ Tesseract.js (offline mode)
Runs entirely inside your browser. No internet required after the first load. No API key. Great for privacy-sensitive content or when you're offline.

── What it's great for ───────────────────

✓ Copy text from images and screenshots
✓ Extract data from tables and charts
✓ Grab code from screencasts or tutorials
✓ Read locked PDFs displayed in the browser
✓ Transcribe handwritten notes
✓ Pull contact info from photos
✓ Digitize scanned documents
✓ Capture text from videos (pause and capture)

── Privacy first ──────────────────────────

Your API key is stored locally in your browser and never leaves your device except when making a direct call to Anthropic's API. No analytics. No tracking. No servers in between. History is stored locally and you can clear it anytime.

── No subscription. No account. ───────────

SnapText is free to use with Tesseract (offline mode) right away. For Claude Vision, you only need a free Anthropic API key — pay only for what you use, at Anthropic's standard rates (typically fractions of a cent per capture).

Get your free API key at console.anthropic.com

────────────────────────────────────────

Try it on any page in 30 seconds. Click "Add to Chrome" to get started.

What's new?
v1.1.0 — Complete rewrite with 99-test coverage. Fixed capture flow reliability, added re-injection guard for repeated captures, improved error messages, and validated all OCR pipelines. Tesseract offline mode now works fully CSP-compliant with locally bundled libraries.

screenshot to text
OCR extension
image to text
Claude Vision
copy text from image
text extractor
screenshot OCR
Tesseract
offline OCR
screen capture text