Choose Val if
- You already live in Cursor or Claude Code and want QA as an MCP tool.
- You want axe-core and Lighthouse in the same agent loop that edits the code.
- You are fine bringing your own model spend on top of $4.99/mo.
Comparison · Val
First published · Facts checked · Updated
Choose Val if you want a $4.99/mo MCP server your coding agent drives against localhost or a live URL, with axe-core, Lighthouse and a fix loop on your own model key. Choose Sybqa if you want a hosted reviewed plan and evidence report without installing an MCP server or bringing a model. We have not run Val.
Val (by Nyx) is a host-driven MCP QA agent: your coding assistant reasons, Val drives Chromium, runs axe-core and Lighthouse, and can loop detect → fix → verify inside the editor. Sybqa is a hosted checking service with a plan you review and an evidence report; it does not live inside Cursor as an MCP server today for this loop. We make Sybqa, so weigh this page accordingly.
Have a site ready to check? Test my site free. Add your URL and review a test plan before starting a run. See a sample report first.
| Point | Sybqa | Val |
|---|---|---|
| Core job | Explores approved journeys for bugs, UI inconsistencies and UX flow issues using manual-testing techniques and real browser checks | MCP QA agent: Chromium browser, structured findings, fix loop via your coding agent |
| Where you start | A link and an optional goal. You review the plan before anything runs | MCP config in Cursor, Claude Code, Codex or Windsurf, or a CLI / GitHub Action |
| What you get back | An evidence report (HTML, JSON and JUnit) with available screenshots and recorded steps; blocked and not-run checks are listed, never counted as passed | Structured review (green/yellow/red), axe and Lighthouse results, PR comments when targeted at a PR |
| Public pricing | A free plan after signup (10 runs a month, no AI review) and a paid plan with AI credits | $4.99/mo subscription; bring your own model key (100% MCP) |
| Browsers | Chromium-based checks at desktop, phone and narrow widths | Chromium only; Firefox/Safari “not yet” per the product page |
| Monitoring | On-demand runs; continuous production monitoring is not the product | On demand when the agent asks; not 24/7 monitoring |
Val’s page says it refuses to fill a live Stripe card and only auto-fills the 4242 test card in test mode. Sybqa likewise refuses live payment credentials in hosted customer mode.
Val’s home page advertises a $4.99/mo subscription and “bring your own key.” It does not describe an ongoing free hosted tier comparable to Sybqa’s free monthly runs.
Val’s own comparison page says hand-written Playwright is best for a fixed regression suite you maintain, while Val is for exploratory QA driven by your AI with no test code. Sybqa can export a Playwright spec from an approved plan; it does not maintain that file for you.
Explore bugs, UI consistency and user journeys from a reviewed plan. Inspect browser evidence and visual review leads in the report. The free plan gives 10 runs a month without AI review after you sign up. See pricing.
Add your site link here and it will be waiting in the console after you create an account. Then describe one journey to check and review the test plan before any run starts.