Give AI agents
eyes and hands

Local HTTP server for screen capture, mouse control, and keyboard input on Windows. Let Claude see your screen and operate your computer.

$ git clone https://github.com/shameindemgg/eyehands.git
$ cd eyehands && pip install -r requirements.txt
$ python server.py

Choose your tier

Free

$0 -- open source

  • Screen capture (screenshot, live view)
  • Mouse control (move, click, scroll)
  • Keyboard input (key, type_text)
  • OCR text search (/find)
  • Background frame buffer (20fps)
  • Composite actions
  • UI Automation
  • Batch operations
Get started

Pro

$19 -- one-time payment

  • Everything in Free
  • Click-at, click-and-wait
  • Type-and-enter, smooth move
  • Batch action sequences
  • Windows UI Automation
  • Find and click elements by name
  • UI element tree inspection
  • 2 machine activations
Buy Pro

API endpoints

MethodEndpointDescriptionTier
GET/pingHealth checkFree
GET/screenshotJPEG screen captureFree
GET/latestLatest buffered frameFree
GET/viewLive HTML viewerFree
GET/findOCR text searchFree
POST/moveRelative mouse moveFree
POST/clickMouse clickFree
POST/keyKeypress with modifiersFree
POST/type_textType unicode textFree
POST/scrollScroll wheelFree
POST/click_atMove + click in one callPro
POST/click_and_waitClick + wait for changePro
POST/type_and_enterType + press EnterPro
POST/batchExecute action sequencePro
POST/smooth_moveSmooth mouse movementPro
GET/ui/windowsList all windowsPro
GET/ui/findSearch UI elementsPro
POST/ui/click_elementFind + click by namePro