Screen

Screen capture and OCR: take screenshots (full, region, window), extract text from images with tesseract.js, list displays, and compare images

Package: @robinpath/screen | Category: Web | Type: Utility

Authentication

No authentication required. All functions are available immediately.

Use Cases

Use the screen module when you need to:

• Take a full screenshot and save to file -- Use screen.capture to perform this operation
• Capture a specific rectangular region of the screen -- Use screen.captureRegion to perform this operation
• Capture a specific window by its title (falls back to full screen if not found) -- Use screen.captureWindow to perform this operation
• Extract text from an image using OCR (tesseract.js) -- Use screen.ocr to perform this operation
• Extract text from a rectangular region of an image -- Use screen.ocrRegion to perform this operation

Quick Reference

Function	Description	Returns
capture	Take a full screenshot and save to file	{path, size}
captureRegion	Capture a specific rectangular region of the screen	{path, size, region: {x, y, width, height}}
captureWindow	Capture a specific window by its title (falls back to full screen if not found)	{path, size, window, fallback}
ocr	Extract text from an image using OCR (tesseract.js)	{text, confidence, language}
ocrRegion	Extract text from a rectangular region of an image	{text, confidence, region: {x, y, width, height}}
setLanguage	Set the default OCR language (eng, rus, deu, fra, spa, chi_sim, jpn, kor, ron, etc.)	{language, previous}
listDisplays	List all available displays/monitors	Array of {id, name, width, height, primary}
compare	Compare two images byte-by-byte to check if they are identical	{identical, image1Size, image2Size, sameSize}

Functions

capture

Take a full screenshot and save to file

Module: screen | Returns: object -- {path, size}

screen.capture "./screenshot.png"

Parameter	Type	Required	Description
filePath	string	Yes	Output file path
options	object	No	{format?: 'png'

---

captureRegion

Capture a specific rectangular region of the screen

Module: screen | Returns: object -- {path, size, region: {x, y, width, height}}

screen.captureRegion "./region.png" 100 100 400 300

Parameter	Type	Required	Description
filePath	string	Yes	Output file path
x	number	Yes	Left position in pixels
y	number	Yes	Top position in pixels
width	number	Yes	Region width in pixels
height	number	Yes	Region height in pixels

---

captureWindow

Capture a specific window by its title (falls back to full screen if not found)

Module: screen | Returns: object -- {path, size, window, fallback}

screen.captureWindow "./notepad.png" "Untitled - Notepad"

Parameter	Type	Required	Description
filePath	string	Yes	Output file path
windowTitle	string	Yes	Exact window title to capture

---

ocr

Extract text from an image using OCR (tesseract.js)

Module: screen | Returns: object -- {text, confidence, language}

screen.ocr "./screenshot.png"

Parameter	Type	Required	Description
imagePath	string	Yes	Path to the image file
options	object	No	{language?: string} e.g. 'eng', 'rus', 'deu'

---

ocrRegion

Extract text from a rectangular region of an image

Module: screen | Returns: object -- {text, confidence, region: {x, y, width, height}}

screen.ocrRegion "./screenshot.png" 50 50 200 100

Parameter	Type	Required	Description
imagePath	string	Yes	Path to the image file
x	number	Yes	Left position in pixels
y	number	Yes	Top position in pixels
width	number	Yes	Region width in pixels
height	number	Yes	Region height in pixels
options	object	No	{language?: string}

---

setLanguage

Set the default OCR language (eng, rus, deu, fra, spa, chi_sim, jpn, kor, ron, etc.)

Module: screen | Returns: object -- {language, previous}

screen.setLanguage "deu"

Parameter	Type	Required	Description
language	string	Yes	Tesseract language code

---

listDisplays

List all available displays/monitors

Module: screen | Returns: object -- Array of {id, name, width, height, primary}

screen.listDisplays

Parameter	Type	Required	Description
(none)		No	Call with no arguments

---

compare

Compare two images byte-by-byte to check if they are identical

Module: screen | Returns: object -- {identical, image1Size, image2Size, sameSize}

screen.compare "./before.png" "./after.png"

Parameter	Type	Required	Description
image1Path	string	Yes	Path to first image
image2Path	string	Yes	Path to second image

---

Error Handling

All functions throw on failure. Common errors:

Error	Cause
(standard errors)	Check function parameters and authentication

@desc "Capture and validate result"
do
  set $result as screen.capture "./screenshot.png"
  if $result != null
    print "Success"
  else
    print "No result"
  end
enddo

Recipes

1. List and iterate Displays

Retrieve all items and loop through them.

@desc "List displays and iterate results"
do
  set $result as screen.listDisplays
  each $item in $result
    print $item
  end
enddo

2. Multi-step Screen workflow

Chain multiple screen operations together.

@desc "Capture, capture region, and more"
do
  set $r_capture as screen.capture "./screenshot.png"
  set $r_captureRegion as screen.captureRegion "./region.png" 100 100 400 300
  set $r_captureWindow as screen.captureWindow "./notepad.png" "Untitled - Notepad"
  print "All operations complete"
enddo

3. Safe capture with validation

Check results before proceeding.

@desc "Capture and validate result"
do
  set $result as screen.capture "./screenshot.png"
  if $result != null
    print "Success: " + $result
  else
    print "Operation returned no data"
  end
enddo

Related Modules

• json -- JSON module for complementary functionality