E2E Check

What it does

Run E2E tests or perform interactive browser-based verification using Playwright. The skill supports two modes: a Runner mode that executes the project's Playwright test suite and parses results (total/passed/failed/skipped, durations, failure summaries), and an Interactive mode that drives a browser for manual walkthroughs, snapshots, and screenshots for visual verification.

When to use it

Use this skill when you need to: run the project's end-to-end tests from CI or locally, validate a failing test with detailed output and screenshots, or manually verify critical user journeys (login, dashboard, reports) via an interactive browser session. It is intended for devs running local servers and test suites (reads dev port from CLAUDE.md or package.json).

What's included

• Scripts: none detected in the skill directory (has_scripts=false) — the skill assumes existing Playwright test commands in the project (package.json or CLAUDE.md).
• References: none included.
• Instructions: mode detection, pre-flight server checks (curl localhost:{port}), running test commands, parsing Playwright output for pass/fail counts and failures, taking/locating screenshots, and interactive step-by-step browser navigation using MCP Playwright helper calls.

Compatible agents

This skill is authored for code-focused agent runtimes (Claude Code / Codex-like tooling) and developer-oriented CLIs that can run Playwright and shell commands.