Playwright Explore Website
Explore a live web app with Playwright MCP before you write E2E tests, so you capture real flows, locators, and test ideas.
Overview
Playwright Explore Website is an agent skill most often used in Ship (also Build) that explores a live site with Playwright MCP and turns findings into locator notes and test-case proposals.
Install
npx skills add https://github.com/github/awesome-copilot --skill playwright-explore-websiteWhat is this skill?
- Navigates a user-supplied URL via Playwright MCP and stops to ask if none is given
- Touches 3–5 core features or user flows with documented interactions
- Records UI elements, locators, and expected outcomes from each step
- Closes the browser context and delivers a concise exploration summary
- Proposes and generates test cases from the exploration findings
- Explores 3–5 core features or user flows per session
- Six-step procedure ending with test-case proposals
Adoption & trust: 10.9k installs on skills.sh; 34.6k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need E2E coverage but do not know which flows, selectors, or outcomes matter on a site you have not mapped yet.
Who is it for?
Solo builders validating a staging URL or new feature area before investing in full Playwright automation.
Skip if: Teams that only test APIs, already have exhaustive E2E specs, or cannot use browser/MCP tooling in their agent environment.
When should I use this skill?
You need website exploration for testing using Playwright MCP and have (or can obtain) a target URL.
What do I get? / Deliverables
After a short MCP-driven walkthrough you get documented interactions, locators, a summary, and draft test cases you can implement in your Playwright suite.
- Exploration summary with interactions and locators
- Proposed test cases based on documented flows
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Ship/testing because the skill’s end goal is exploration-driven test design, even though you often run it right after a build milestone. Subphase testing fits guided site walkthroughs, UI element documentation, and proposed test cases—not one-off debugging.
Where it fits
Walk a freshly built dashboard to see which nav paths and forms need regression tests.
Explore checkout and auth on staging before codifying Playwright specs for release.
Re-explore production-like UI after a redesign to refresh broken locator assumptions.
How it compares
Use for agent-guided discovery and test ideation instead of jumping straight into unguided Playwright script generation.
Common Questions / FAQ
Who is playwright-explore-website for?
Indie and solo builders who use Claude Code, Cursor, or similar agents with Playwright MCP and want structured site exploration before writing tests.
When should I use playwright-explore-website?
During Ship/testing when you have a URL and need flow maps and locators; also right after Build when a new UI ships and coverage is thin.
Is playwright-explore-website safe to install?
It drives real browser navigation via MCP; review the Security Audits panel on this Prism page and only point it at URLs you trust.
SKILL.md
READMESKILL.md - Playwright Explore Website
# Website Exploration for Testing Your goal is to explore the website and identify key functionalities. ## Specific Instructions 1. Navigate to the provided URL using the Playwright MCP Server. If no URL is provided, ask the user to provide one. 2. Identify and interact with 3-5 core features or user flows. 3. Document the user interactions, relevant UI elements (and their locators), and the expected outcomes. 4. Close the browser context upon completion. 5. Provide a concise summary of your findings. 6. Propose and generate test cases based on the exploration.