Overview

What It Helps You Do

Use this skill to convert acceptance criteria into Playwright test automation. It validates whether your AC are automation-ready, then generates validated test plans and executable test suites.

Activate it with:

• /accelint-ac-to-playwright
• Phrases like "review these AC" or "convert these AC to tests"
• Related requests about Playwright-ready acceptance criteria, Gherkin conversion, or test generation

Three main workflows:

• Assess readiness: Check if AC follow automation patterns before writing tests
• Convert to tests: Transform .md bullets or .feature Gherkin files into Playwright specs
• Fix validation gaps: Get specific feedback on what needs to change (missing targets, vague actions, unmeasurable outcomes)

When to Use

Use this skill when users ask to:

• Review, evaluate, or check if AC are ready for automation
• Assess if AC can be converted as-is
• Validate AC quality for Playwright
• Turn AC into tests or generate tests from acceptance criteria
• Convert .md bullets or .feature Gherkin files to Playwright specs
• Create test automation from requirements

How It Works

The skill operates in two modes:

Assessment Mode

Analyzes AC text without generating files. Validates:

• Structure: Proper bullet markers or Gherkin syntax, correct step ordering (Given → When → Then)
• Targets: Complete area.component.intent pattern using controlled vocabulary
• Actions: Recognized verbs (click, fill, select) with literal values where needed
• Outcomes: Explicit, measurable assertions

Reports either "✓ AC are conversion-ready" with checklist or "❌ Issues found" with specific problems, examples, and fixes.

Conversion Mode

Generates complete test automation:

1. Runs assessment validation first (stops if issues found)
2. Generates JSON test plan using area.component.intent selectors
3. Validates plan against schema (max 2 attempts)
4. Translates plan to Playwright spec file
5. Creates summary metadata and copies shared fixtures

Target pattern: Each action specifies area.component.intent for predictable, maintainable selectors—area identifies location, component identifies UI element type, intent identifies purpose.

Output mapping: One AC file → one test suite. Markdown bullets become individual tests; Gherkin scenarios become tests, with Scenario Outline rows becoming parameterized cases.

Prerequisites

• Playwright installed with TypeScript support
• AC files in bullet markdown (.md) or Gherkin (.feature) format
• For validation: Skill dependencies built (npm run build in skill directory)

What You Get

Assessment mode:

• Validation report with issues by file (problem, example, fix) or confirmation that AC are ready

Conversion mode:

• JSON test plan (<plans-dir>/<suite-slug>.json)
• Playwright test suite (<tests-dir>/<suite-slug>.spec.ts)
• Summary metadata (JSON batch file)
• Shared fixtures/ directory (copy both .spec.ts files and fixtures to your project)

Examples

Example: Assessing AC Quality

/accelint-ac-to-playwright "Review login-flow.feature and check if it's automation-ready"

# Output:
❌ AC are not conversion-ready. Issues found:

File: login-flow.feature
1. Scenario "User logs in": Missing target intent
   - Problem: Selector incomplete (button.form vs button.form.submit)
   - Example: "clicks the button on the form"
   - Fix: Specify intent: "clicks the Submit button on the form"

Example: Converting AC to Tests

/accelint-ac-to-playwright "Convert form-validation.feature to Playwright tests.
Plans in ./test-plans, tests in ./e2e/tests, summaries in ./test-summaries"

# Generates:
# - ./test-plans/form-validation.json
# - ./e2e/tests/form-validation.spec.ts
# - ./test-summaries/batch-[timestamp].json
# - ./e2e/tests/fixtures/

Good to Know

Good to know: Assessment mode analyzes without writing files—use "review" language for validation, "convert" language for generation.

Good to know: The skill asks for explicit output directories before writing artifacts.

Good to know: Targets need the full area.component.intent pattern. Incomplete selectors like button.form are too ambiguous; button.form.submit is automation-ready.

Good to know: AC must use specific action verbs (click, fill, select) with literal values like 'premium@example.com' rather than placeholders like "a valid email". Vague verbs like "interact with" cannot map to Playwright actions.

Good to know: The skill only generates assertions for outcomes stated explicitly in AC. Omitted expected behavior results in omitted assertions.

Good to know: Validation stops after 2 failed attempts rather than guessing at corrections. Plans must be structurally valid before tests are generated.