How does the multi-agent system work?

The skill uses a 'test-lead' agent to coordinate five sub-agents—explorer, functional, visual, API, and performance—who work in parallel to analyze different aspects of your site.

What kind of output does this skill provide?

It generates a detailed directory containing a master REPORT.md, individual JSON reports for each testing category, and screenshots for mobile, tablet, and desktop layouts.

What is required to run the playwright-test skill?

You must have the Playwright MCP server (@anthropic/mcp-playwright) installed and configured within your Claude Code settings for the agents to control the browser.

Does it support testing of local development environments?

Yes, you can run tests against local URLs like http://localhost:3000 as long as the server is running and accessible to the browser instance.

Can it handle single-page applications (SPAs)?

Yes, the page-explorer agent is designed to detect SPA frameworks, dynamic content, and complex routing during the initial discovery phase.

Playwright AI Test Team

Name: Playwright AI Test Team
Author: intenet1001-commits

byintenet1001-commits

0•

보안 및 테스팅

Orchestrates a multi-agent AI team to perform comprehensive end-to-end web testing, accessibility audits, and performance analysis.

This skill deploys a specialized team of six AI agents to conduct deep-dive testing of web applications using Playwright. It automates the entire QA lifecycle—from initial page exploration and structural mapping to parallelized functional testing, visual regression, accessibility checks, and API traffic analysis. By monitoring Core Web Vitals and capturing responsive screenshots, the skill culminates in a professional, unified Markdown report that prioritizes issues and suggests optimizations, making it an essential tool for maintaining high-quality web standards in any development environment.

주요 기능

010 GitHub stars

02In-depth performance auditing including FCP, LCP, CLS, and TTFB metrics.

03Real-time API interception and network traffic analysis to detect backend failures.

04Automated page mapping and structure analysis of complex SPAs and multi-page apps.

05Multi-agent orchestration with six specialized roles for parallelized deep testing.

06Comprehensive visual inspections across mobile, tablet, and desktop viewports.

사용 사례

01Debugging complex UI interactions and API race conditions in a staging environment.

02Running automated regression tests and accessibility audits before a production release.

03Generating a comprehensive site health report for a new or inherited web project.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add intenet1001-commits/cs_plugins playwright-test

For use in Claude.ai and ChatGPT

Download Skill