OCR & Screenshots

OCR extracts text from images and PDFs using optical character recognition. Pass a URL or a namespace file path. An optional language hint improves recognition accuracy for non-English content.

OCR

From a URL

Example

OCR "https://example.com/receipt.png" SET ?text
AFTER EMIT ?text

From a file

The AS modifier accepts an ISO 639-2 language code. When omitted, English ("eng") is assumed.

Example

OCR "/root/files/invoice.png" AS "eng" SET ?text
AFTER EMIT ?text

Screenshots

Capturing a real screenshot of the visitor’s screen requires their explicit, unavoidable consent via the browser’s own native picker. This is documented in full, alongside every other device permission, on USER DEVICE & Syncing.

Capturing a Screenshot

USER DEVICE SCREENSHOT LABEL "Share your screen to continue" SET ?device
AFTER EMIT ?device("screenshot")