Desktop infrastructure for AI agents.
npm install orgo
# or
yarn add orgo
# or
pnpm add orgo
import { Computer } from 'orgo';
// Create computer
const computer = await Computer.create();
// Control
await computer.leftClick(100, 200);
await computer.type("Hello world");
await computer.key("Enter");
await computer.screenshot(); // Returns Buffer
// Execute code
await computer.exec("import pyautogui; pyautogui.click(512, 384)");
// Cleanup
await computer.shutdown();
const computer = await Computer.create({
apiKey: 'your-api-key', // or set ORGO_API_KEY env var
projectId: 'existing-project-id', // optional
config: { /* custom config */ }, // optional
baseApiUrl: 'https://custom-api.com' // optional
});
-
leftClick(x, y)
- Left mouse click -
rightClick(x, y)
- Right mouse click -
doubleClick(x, y)
- Double click -
scroll(direction, amount)
- Scroll in a direction
-
type(text)
- Type text -
key(key)
- Press key or key combination
-
screenshot()
- Get screenshot as Buffer -
screenshotBase64()
- Get screenshot as base64 string
-
exec(code, timeout?)
- Execute Python code -
bash(command)
- Execute bash command -
wait(seconds)
- Wait for specified seconds
-
prompt(options)
- Control computer with Claude 4
-
status()
- Get computer status -
restart()
- Restart the computer -
shutdown()
- Terminate the computer
Full documentation: docs.orgo.ai
This package includes TypeScript declarations. All types are exported and available:
import type { ComputerConfig, ComputerStatus, PromptOptions } from 'orgo';
MIT