Browser Automation
Automates real browser interactions to validate UI behavior, capture screenshots, and perform end-to-end testing within Claude Code.
This skill enables Claude to interact with actual web browsers to verify rendered UI states, debug frontend issues, and generate visual evidence like screenshots or recordings. It intelligently selects the best runtime—prioritizing platform-specific tools or project-specific runners like Playwright—to execute deterministic flows, handle authentication, and ensure accessibility compliance. By focusing on semantic targets like roles and labels rather than brittle CSS selectors, it provides a robust way to explore and validate web applications during the development process.
Key Features
01Automated screenshot and recording capture
02Rendered UI exploration and validation
03Semantic targeting using roles, labels, and test IDs
04Smart runtime selection (Built-in, Project Runner, or Playwright)
05End-to-end flow execution and test generation
0633 GitHub stars
Use Cases
01Generating and running automated end-to-end tests for feature verification
02Debugging complex frontend behaviors that require a rendered DOM environment
03Auditing web accessibility and visual regressions using snapshots