Browser Automation Safety
Enforces secure and efficient browser interactions for UI validation, web scraping, and automated testing within agent workflows.
This skill provides a robust framework for managing browser automation, emphasizing security and reliability during web-based tasks. It guides the agent through identifying existing capabilities like Playwright, navigating safely within user-defined origins, and using semantic selectors to ensure stable UI interactions. By implementing strict rules against auto-installing packages and handling sensitive authentication data via file paths rather than raw strings, it mitigates common risks associated with autonomous browser control while ensuring all sessions are properly terminated.
Key Features
01Secure authentication handling using file-based states and cookie exports
02Strict safety boundaries to prevent unauthorized navigation or prompt injection
03Automatic detection of Playwright and local development servers
04Semantic selector prioritization for stable and accessible automation
05Comprehensive cleanup protocols for browser processes and temporary artifacts
067 GitHub stars
Use Cases
01Scraping data from authenticated web sessions without exposing secrets
02Capturing redacted screenshots and console logs for debugging evidence
03Validating rendered UI components and local dev server outputs