claude-qa-toolkit

Full QA audit orchestrator. Use when the user asks to "test my app", "run a full QA audit", "test everything", or invokes /qa-audit. Discovers what kind of application this is, selects the relevant QA agents (ui-inspector, api-sentinel, security-scout, perf-guardian, data-validator, regression-watcher), runs them, and merges their reports into one master QA report.

Onboard the QA toolkit onto any project. Use on first contact with a new codebase, when the user says "set up QA", "qa init", "/qa-init", or when any QA agent finds no qa-profile.md. Studies the project deeply and writes the persistent QA profile that every other agent reads first — the toolkit's long-term memory for this project.

Wire the QA suite into CI. Use when the user wants tests running automatically on every PR or on a schedule: "add this to CI", "run tests on every push", "/qa-ci". Generates GitHub Actions workflows for the generated Playwright suite and (optionally) scheduled full audits.

Self-healing for the generated test suite. Use when previously-green Playwright tests are failing, the user says "my tests broke", "fix the tests", "/qa-heal", or after a UI refactor breaks selectors. Classifies each failure (real bug vs test drift) and repairs the drift — never papers over real bugs.

Compare two QA audit runs and report what improved, what regressed, and what persists. Use when the user asks "did it improve since last time?", "compare audits", "QA trend", or invokes /qa-trend. Requires at least two timestamped runs under qa-reports/runs/.

Audit web pages for WCAG 2.1 compliance. Use when the user asks about accessibility, a11y, WCAG, screen readers, or contrast issues.

Generate and execute API contract tests from OpenAPI/Swagger specs. Use when the user wants contract tests or schema validation for an API.

Transform informal bug descriptions into structured, professional bug reports. Use when the user describes a bug and needs it documented or filed.

Turn raw CI/CD pipeline output into a readable report. Use when the user pastes build/pipeline logs or asks to summarize CI results.

Identify and fix intermittently failing tests. Use when the user mentions flaky tests, random failures, or unreliable CI.

Generate and run load test scripts for APIs and pages. Use when the user asks for load tests, stress tests, or traffic simulation scripts.

Parse application logs and find root causes. Use when the user shares logs, stack traces, or asks what is causing errors.

Generate SQL queries that validate data quality and integrity. Use when the user wants validation queries for a database.

Generate realistic test data (SQL, CSV, JSON) for any schema. Use when the user needs seed data, fixtures, or fake records.

Generate structured test plans from requirements or user stories. Use when the user asks for a test plan, test cases, or coverage matrix.