Brings Playwright browser automation into Claude and other AI editors through 13 MCP tools. You point it at a codebase and a running app, and it scans your source to understand real routes and form fields, then opens a headless browser to explore pages, fill forms, and click through flows. It generates test scenarios based on code coverage gaps, executes them step by step with screenshots, and produces HTML reports showing what broke. Handles authentication, element scanning, and iterative test convergence until you hit pass rate thresholds. Useful when you want AI to drive end-to-end testing without writing Playwright scripts yourself, especially for catching regressions during rapid prototyping.
claude mcp add --transport stdio aishwaryshrivastav-vibe-testing -- npx -y vibe-testing