QA

Use when a page or flow needs a WCAG 2.1 AA conformance audit. Runs axe-core against the live DOM via Chrome DevTools MCP, inspects the accessibility tree, verifies keyboard navigation and focus order, checks color contrast, ARIA roles, landmarks, headings, form labels, skip links, and alt text — and reports every violation with a code-level fix suggestion.

Use when an OpenAPI 3.x specification (or URL to one) needs to be turned into a runnable API test suite. Generates Schemathesis property-based tests, REST-assured (Java) suites, or supertest (Node) specs covering positive, negative, auth, and rate-limit flows — and validates every response against the declared schema.

Use when a defect needs to be filed in a structured, reproducible format. Produces a bug report with an observed-behavior title, environment, numbered steps to reproduce, expected vs. actual, frequency, severity, priority, attachments, related tickets, and workaround. Enforces reproducibility before filing.

Use when a user story, acceptance criteria, or Jira ticket needs to be turned into a runnable Playwright or Cypress end-to-end test. Generates page-object-model specs with data-testid selectors, explicit waits, network mocking for edge cases, screenshot-on-failure, and inline accessibility assertions.

Use when a page, release candidate, or route needs a Core Web Vitals and Lighthouse audit with prioritized fix recommendations. Runs Lighthouse via Chrome DevTools MCP, captures a performance trace, analyzes long tasks and render-blocking resources, and produces actionable guidance per metric (LCP, INP, CLS, TTFB, FCP).

Use when realistic, seeded test data is needed from a JSON Schema, OpenAPI component schema, or database schema (PostgreSQL / MySQL / SQLite DDL). Generates reproducible records that honor constraints (unique, min/max, enum, foreign keys, formats) using faker libraries, and emits SQL inserts, JSON, or CSV. Supports a seed for deterministic runs.

Use when preparing a release build for production and a complete pre-release QA sweep is required. Orchestrates API, E2E, performance, and accessibility testing, then consolidates results into a single release-readiness report with a go/no-go recommendation.