xyva.ai - QA Studio
Local-first QA orchestration. Swarm QA, accessibility audits, and a Playwright test runner - entirely on your machine, BYOK for every AI provider.
Features at a Glance
| Feature | Availability | Description | |
|---|---|---|---|
| Wizard | Guided Onboarding | Solo + Team | Step-based setup with project scan, provider setup, and preflight checks |
| Test Hub | Test Catalog | Solo + Team | Centralised test catalog with data-driven parameterisation and dataset management |
| Runner | Playwright Execution | Solo + Team | Playwright-based runner with parallel execution, debug mode, and CI integration |
| AI Assistant | Project-aware chat | Solo + Team | Conversational AI with full project context - code generation, reviews, and assistive suggestions |
| Test Builder | Test scaffolding | Solo + Team | Generate Playwright tests from a URL or description as a starting point |
| Swarm QA | 8 focused agents | Solo + Team | Link Patrol, HTTP Guard, A11y Scout, Smoke Flow, Perf Sentinel, SEO Recon, Form Fuzzer, API Health |
| Accessibility Auditor | BFSG / WCAG 2.1 AA | Solo + Team | Axe-based audits with evidence reports |
| Architecture Audit | Dependency insight | Team | Project structure, dependency graph, component analysis |
| Evidence & Reports | Screenshots & traces | Solo + Team | PDF export, visual regression, structured run reports |
| Ops Portal | Infrastructure health | On request | Private deployment and advanced operations visibility |
| Swarm Schedule | Automated scans | On request | Recurring Swarm QA with webhook-based escalation for private rollouts |
Swarm QA - 8 Specialised Agents
| Agent | What it does |
|---|---|
| Link Patrol | Crawls all links, checks external domains, detects redirect chains |
| HTTP Guard | Monitors HTTP errors, network failures, console errors |
| A11y Scout | WCAG/BFSG accessibility checks - alt text, ARIA, contrast, landmarks |
| Smoke Flow | Page-load verification, JS error detection, rendering failures |
| Perf Sentinel | Core Web Vitals (LCP, CLS, TTFB), resource sizes, load benchmarks |
| SEO Recon | Meta tags, Open Graph, canonical URLs, structured data, sitemap |
| Form Fuzzer | XSS probes, SQL injection, overflow, empty submit, error feedback |
| API Health | Endpoint availability, response time, status code validation |
All agents run in parallel. Findings are classified by severity (critical -> info) and action category (fix-now -> ignore).
Quick Links
Getting Started
Core Features
AI
Compliance & Audit
Private Deployment & Ops
Licensing