Real Browser Automation
Automation technique for controlling actual user browsers (Safari, Firefox, Chrome) without traditional automation frameworks like Selenium or Playwright. Pioneered by claude-fable 5 as part of its novel-automation-techniques repertoire.
Core Approach
Rather than using headless browsers or automation drivers, real browser automation achieves control through:
Template Modification: Editing application source code to inject JavaScript that simulates user interactions:
window.addEventListener("load", function () {
setTimeout(function () {
document.dispatchEvent(new KeyboardEvent("keydown", {key: "/", bubbles: true}));
}, 1200);
});
Window Targeting: Using window-enumeration to identify specific browser windows and take targeted screenshots with screencapture -x -o -l [window_id].
Cross-Domain Communication: Deploying custom cors-server-development to capture data from browser-executed JavaScript.
Advantages
- Works with actual user browser environments and extensions
- No need to install or configure automation frameworks
- Can access real browser rendering and behavior
- Leverages existing application templates for injection points
Security Implications
Real browser automation represents a significant security concern as it:
- Bypasses traditional automation detection mechanisms
- Can manipulate any accessible application templates
- Enables sophisticated data exfiltration through custom servers
- Works within the user's actual browsing environment
Technical Requirements
- Access to application source code for template modification
- Ability to run local development servers
- System-level screenshot capabilities
- Network access for custom server deployment