playwright-explore-website

Explores websites via Playwright MCP to identify features and generate test cases.

Updated Aug 28, 2026
One-click install
npx skills add https://github.com/miyake-san/sogo-agent-platform --skill playwright-explore-website-miyake-san
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: playwright-explore-website
Source: https://github.com/miyake-san/sogo-agent-platform/tree/main/skills/core/playwright-explore-website
Command: npx skills add https://github.com/miyake-san/sogo-agent-platform --skill playwright-explore-website-miyake-san

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @playwright/mcp.

What problem does it solve? Manually exploring a website to understand its core functionality before writing tests is slow and inconsistent. This Skill automates structured website exploration through the Playwright MCP server, documenting user flows and producing test cases from real browser interactions. ## Core Features & Use Cases - Guided Website Exploration: Navigates to a provided URL and interacts with 3-5 core features or user flows using the Playwright MCP server. - Interaction Documentation: Records user interactions, relevant UI elements with their locators, and expected outcomes for each flow. - Test Case Generation: Proposes and generates test cases based on the documented exploration findings. - Use Case: Point the Skill at a new web application under development; it explores the login, search, and checkout flows, documents the element locators, and outputs a set of candidate test cases for your QA suite. ## Quick Start Explore the website at https://example.com using Playwright, document its core user flows, and generate test cases based on your findings.

Frequently Asked Questions about playwright-explore-website

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I explore a website for testing with Playwright MCP?

Provide a target URL and the Skill navigates to it using the Playwright MCP server, interacts with 3-5 core features or user flows, and documents each interaction. It then summarizes findings and proposes test cases based on the exploration.

How to generate test cases from website exploration?

The Skill documents user interactions, UI element locators, and expected outcomes during exploration, then proposes and generates test cases from those findings. This turns manual exploratory testing into structured, reusable test documentation.

What is required to run Playwright MCP website exploration?

You need the Playwright MCP server, installable via npx @playwright/mcp@latest or from the microsoft/playwright-mcp GitHub repository. The MCP server drives the browser so the Skill can navigate and interact with pages.

Does the exploration handle browser cleanup automatically?

Yes, the Skill closes the browser context upon completion of the exploration. This ensures no lingering browser sessions remain after the findings and test cases have been produced.

What are the limitations of automated website exploration for testing?

The exploration covers 3-5 core features or flows per run, so it does not exhaustively map large applications. It also requires a reachable URL and a working Playwright MCP server, and complex authenticated flows may need manual guidance.