test_skill

This skill helps you run end-to-end tests with Playwright MCP in interactive mode and generate markdown test reports under docs/testing.

0

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill eljun/workflow-skills --skill test

SKILL.md36.3 KB

Overview

This skill tests web implementations using Playwright MCP for interactive E2E debugging or the Playwright test runner for CI. It runs flows, verifies email-based auth via Mailinator, and writes final reports to docs/testing/{task-name}.md. Use interactive mode for guided browser inspections and CI mode for headless, repeatable test runs.

How this skill works

The skill detects mode first: interactive (requires Playwright MCP tools) or CI (--ci flag). Interactive mode drives the browser through mcp__playwright__browser_* tools without creating test files. CI mode generates tests under tests/e2e/{task-name}/, runs npx playwright test, captures results, and optionally cleans artifacts. Email flows use Mailinator to register, verify, and test password resets and notifications.

When to use it

Validate a feature marked "Testing" in TASKS.md after implementation
Run visual, step-by-step verification and debugging (interactive mode)
Execute headless regression suites in CI pipelines (--ci)
Verify email-based authentication, verification, and password reset flows
Generate a single authoritative test report in docs/testing/{task-name}.md

Best practices

Run mode detection first: never switch modes automatically; follow the --ci flag
Use interactive mode only when Playwright MCP tools (mcp__playwright__browser_*) are available
Keep test scripts under tests/e2e/{task-name}/ in CI mode; never place scripts in project root
Always write the final report to docs/testing/{task-name}.md and nothing else in docs/testing/
Ask whether test scripts should be retained or cleaned when running CI (--ci vs --ci --cleanup)

Example use cases

Interactive debugging of a new signup flow: step through pages, take MCP screenshots, confirm Mailinator verification links
CI regression run for payment or dashboard flows, keep scripts for future PRs using /test --ci {task}
One-off verification of a minor bug fix with cleanup: /test --ci --cleanup {task} to remove test artifacts after run
Password reset flow validation: trigger reset, confirm Mailinator email, follow link, assert new password works
Smoke test before merging: generate tests in tests/e2e/{task}/, run headless, and publish docs/testing/{task}.md

FAQ

Stop and configure Playwright MCP. Do not fall back to CI or use curl. Add .mcp.json and restart the environment, or run /test --ci {task} for CI mode.

Where must I put the final test report?

Write the final, single report to docs/testing/{task-name}.md. Do not create other test reports or .spec files in docs/testing/.